Archive

Archive for August 12, 2020

Summary of My Post-CPI Tweets (August 2020)

August 12, 2020 4 comments

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!

  • Well, it’s CPI day and I have to tell you I’m looking forward to this one and I’ll tell you why.
  • Used cars! The Black Book retention index jumped about 9% last month after 8.5% the prior month. There’s typically a 3-month lag before it gets into CPI, but w/ a big move it’s harder to say. Each of those jumps would be worth about 0.3% on core, and we have two of them coming.

  • That being said, (a) there is still one dip we haven’t seen yet so we COULD have a dip in used cars this month. It would be surprising, but it would mean we can prep for a couple of really good numbers. So I’m excited either way.
  • And it’s not just used cars. Now that things are opening up, we’re going to see pressures in other places. Medical care started to show some ups last month and I expect that to continue as hospitals are hurting for revenue.
  • Last month we also saw strong apparel, lodging away from home, and airfares, which were rebounding from the covid-induced swoon. I think that could continue, and it’s an interesting story line to watch.
  • On the other hand – this is next month’s story but college tuitions are likely to decline this year because colleges are giving discounts. Even though the product has changed in quality (e-learning not the same as in-person), the BLS has decided it can’t quality-adjust easily.
  • So CPI for college-tuition-and-fees – again, probably next month – will fall and then rebound hard next year. So that’s fun. It’s not right, but it’s okay, as the saying goes.
  • Shelter last month, ex-hotels, was soft. That’s the only fly in the inflationary ointment, but it’s a big one. So far it doesn’t look like rents are likely to decelerate much overall, nor housing prices fall – but if they do, that’s a big deal.
  • I will be looking at core-ex-housing to see if pressures are broadening, but looking at shelter b/c it’s a big, slow item. If shelter weakens appreciably, it will be news – historically, with the last recession being an exception, housing prices and rents almost never FALL.
  • But they can slow, and with incomes sketchy housing inflation probably SHOULD slow. If it doesn’t, that’s a real sign that the rising monetary tide is raising all assets. (And goods and services). FWIW, wages also aren’t slowing. Atlanta Fed wages are +3.8% y/y.
  • (There’s interesting stuff around the disconnect between wages and the unemployment rate right now, but I’ll save that for a blog post another time. Not really a CPI-day thing.)
  • Consensus today is for 0.2% on core CPI, but a soft 0.2% with y/y falling to 1.1%. I think there’s lots of upside to that if Used Cars pops, but a little downside if shelter is weak again. I’m in the “probably higher” camp.
  • Good luck! And if you’re curious about what an inflation guy does when it’s not CPI day, stop by Enduring Investments: http://enduringinvestments.com
  • Oh, yes.
  • I don’t think we need to worry much about the rounding this month. Core +0.6% m/m; y/y to 1.6% when it was expected to drop to 1.1%.
  • FWIW, that was rounded down. +0.62% m/m on core. Repeat: rounded down. I will have to check but that is the biggest monthly figure in decades.
  • I think I soiled myself.
  • There are going to be a lot of crazy charts like this one this month. This is the last 12 core CPI prints.

  • y/y core rose from 1.19% to 1.57%, in one month. Core goods were -1.10% y/y; now they are -0.5%. Core services were 1.90%; now they’re 2.3%.
  • Take that Keynesians. WHERE’S YOUR OUTPUT GAP MODEL NOW? …but I shouldn’t celebrate. All of those degrees…and poor Nomura forecasting outright deflation…
  • Now interestingly, Used Cars and Trucks was up, 2.33% m/m, but that’s not the big jump yet. (!)

  • Lodging Away from Home, another COVID-casualty, was +1.2% m/m. Same as last month. But the y/y is still -13.26% (was -13.92%).
  • Primary rents rebounded some, +0.19% vs +0.12% last month, and OER as well +0.21% from +0.09%. Those are m/m numbers, and the y/y are still softening though: 3.12% for primary rents and 2.80% for OER, down from 3.22%/2.84%. But not collapsing.
  • OK, I said I was going to be interested in core-ex-housing. It jumped from 0.35% y/y to +1.01% y/y. Now, that’s only the highest since March but again: the deflation dragon, if not slain, is pretty sick.
  • Apparel was +1.08% m/m, but y/y is still -6.4% (was -7.2%). Like the other belly-flop categories, there’s still a lot of recovery to come.
  • So how are the doctors doing? Medical Care was +0.41% m/m, but that actually dropped the y/y slightly to 5.02% from 5.08%. However, that’s mostly because Pharma remains weak.
  • CPI for Medicinal Drugs was flat again. +0.02% this month; -0.01% last month. The y/y is down to 1.1%.

  • But Physicians’ Services up to 2.58% y/y (up 0.67% this month)

  • And hospital services hanging out at around 5% y/y. Look, like many services these are all becoming more labor-intensive and that means…more expensive. Some of that might come back, some day.

  • Totally forgot airfares: +5.4% m/m after +2.6% m/m last month. But still down a lot from the peak. Here’s the y/y figure.

  • And a quick check of the markets: 10-year breakevens +3.5bps, kinda surprised it’s not more. 5y breakevens +5bps. Some of this might just be time for price discovery. I know when I was a CPI swaps dealer, it took some time before we knew wth was the right price.
  • BTW that core increase was the biggest monthly increase since 1991. That predates TIPS by 6 years.
  • Now, college tuition and fees rose to 2.09% y/y from 1.74% y/y. That’s interesting, as that serious ought to be declining next month. And for tuitions, that’s a largeish m/m change. Interesting.
  • Let’s see. Biggest m/m declines: misc personal goods (-41.7% annualized) and meats, poultry, fish & eggs (-36.9%, but it had been up a lot too).
  • The list of gains annualizing more than 10% has 14 categories. Includes motor vehicle insurance, car/truck rental, public transportation, used cars/trucks, communication, jewelry, footwear, lodging away from home…
  • Now, those who live by Median CPI ought to also die by Median CPI. I’ll convert you all, eventually. Median this month will be something like +0.21%, because it ignores the upside long tails like it did the downside ones. y/y will actually decline to 2.56%
  • The message there is just that the underlying trends are pretty stable. But it’s not insignificant that the tails shifted to the right side from the left side. As I’ve said before, that’s sort of what infl looks like in practice, just as disinflation has one-offs to the left.
  • Health insurance y/y is a little softer, down to 18.7% y/y from 19.4%. So we got that going for us.

  • This is the distribution of y/y changes in the CPI. There’s still a big left tail anchor which is why core is below median. But this is a much more balanced distribution than it has been in a while.

  • And here’s the weight of categories going up by more than 2.5% y/y. The weight is the highest since July 2008.

  • That doesn’t look very deflationary to me.
  • Putting together the four-pieces charts and then I’ll wrap up.
  • Piece 1 – food and energy. With all of the wild swings, it’s net-net kinda boring.

  • Piece 2, core goods. This was the piece that was getting a wind behind it because of trade frictions when the crisis hit. Big bounce this month. Much of that is autos, but as I pointed out early: the BIG jump in car prices hasn’t hit the data yet.

  • Piece 3, core services less rent of shelter. Also a big recovery, and some of this is airfares. Some also is medical care. But there are a number of other categories contributing here. Still kind of trendless last 5y, overall.

  • Piece 4, the biggest and slowest piece, and looks scary. Until you remember this includes hotels (lodging away from home). If you take that out, shelter has decelerated some but not a lot, and certainly not in a disturbing way like this appears. Don’t project this!

  • OK to sum up. I saw someone call this a “noisy” report. Well, only in the sense of clanging cymbals. The data here all swung in one direction – but there really weren’t a lot of surprises, per se. The only surprise was the synchony of the surprises to one side.
  • As I said up top, we still have a couple of +0.3% boosts (maybe +0.2% if we’re ahead of mode) coming from used cars. And a lot of the beaten-down categories haven’t really recovered (apparel, etc) fully.
  • THAT’S what’s surprising. This wasn’t the left tail snapping back, much. This was a much broader advance than the decline had been. The decline had been 4 categories: lodging AFH, used cars, airfares, apparel. Way more here.
  • There are lots of bumps ahead, including the question of whether home prices and rents decelerate when and if incomes decline. We aren’t seeing that yet. And with M2 growing at 23% per year, it’s hard to believe asset prices can decline very much. Including housing.
  • The fun thing to think about is: what is happening at the Fed today? Are they clapping wildly, that they succeeded in pushing prices up? Or are they somber, wondering if they might have overdone it? Or are they focusing on median CPI and saying, meh?
  • My guess is that there’s a bit of nervousness. The Fed wants to overshoot 2%, but they don’t want to put it at 6%. I’ve said for a long time: creating inflation is easy. Creating A LITTLE inflation is hard.
  • Well, that was fun. Thanks for tuning in. I’ll put a summary on my blog relatively shortly. Again, if today’s number makes you think ‘hey, maybe we should talk to an inflation guy and see if he can help us’, stop by our website: https://enduringinvestments.com Have a nice day.

I don’t know whether to be exhausted or energized. I think I’ll go with energized, because this is not likely to be the last surprise in inflation prints. We are entering a period, not only of higher inflation (probably), but also much higher inflation volatility. That’s important, because a key underpinning of the valuation argument for stocks and bonds is that inflation is not only low, it’s low and stable and therefore can be ignored in calculations. But if inflation is volatile, and especially if it’s high and volatile, then  companies and investors need to include it in their calculus. And if the inflation factor ends up becoming significant again, after more than a decade of irrelevance, then it means that (a) stocks and bonds will become increasingly correlated and (b) stock and bond valuations will be lower.

Now, I don’t know if the markets really understand what’s going on. In fact, this number was so outside of expectations I think that investors just dismissed it as a one-off, like April’s number. But it’s not. This was not just a snapback of the depressed categories; indeed, most of the categories that were depressed because of Covid (lodging away from home, airfares, e.g.) are still depressed although they’ve rebounded a little. This was much broader than that. But investors have pushed 10-year breakevens up only 3-4bps, to 1.66%. Stocks are soaring, and 10-year nominal yields are a mere 3.5bps higher. Commodities are flat. Gold, after a bloodbath yesterday, is flat today. The only way those reactions make sense is if investors are missing the significance.

When the unemployment rate shoots higher, then you can understand a positive market reaction because investors have come to count on the Fed supporting markets in that circumstance. But that reasoning doesn’t make sense here. Nothing about a 7.2% annualized rate of inflation (0.6% * 12) would make the Fed eager to add more liquidity. Ergo, it must be that investors just don’t care about the inflation numbers, or they think this is a random miss.

They should care. This isn’t a one-standard-deviation miss; it’s the biggest monthly print in thirty years and there was no big outlier. While it doesn’t guarantee that inflation is heading higher, the question is whether this print is consistent with our a priori model of the world.

If you’re a Keynesian, the answer is absolutely not. So economists who are output-gap focused are going to say that this number doesn’t matter; it’s ‘quirky’ or ‘noisy’ or ‘measurement error’; the output gap is going to drag down inflation. Maybe that’s why investors are nonchalant about this…because they’re being told by the bow-tie set to look through it.

But if you’re a monetarist, this is entirely consistent with your a priori model. The only surprising thing about this is that it is happening so soon. I was thinking we would see inflation rise starting in Q4 and it would get messy in 2021. I might have to move up the timetable. Because this number is entirely consistent with my model, I’m much less sanguine. This might be only the first shot over the bow… Indeed, over the next several months I can say that since we are confident that used car prices are going to add a lot to core inflation, we will probably have at least one or two more prints of 0.4%-0.5% on core over the next three months. If we get 0.4%, 0.2%, 0.4%, then in three months core CPI will be back to 2.25% y/y, and that with unemployment still in the high-single or low-double-digits. And it could actually be worse than that. Without home prices collapsing, it’s hard to see it being much better than that and absolutely no way to see how prices (or even the inflation rate) could be lower than that unless something really, really weird happens.

Well, 2020 is the year of really, really weird so I suppose I will never say never. But inflation hedges remain super cheap; if you’ve been waiting to scoop them up I can’t see any argument for waiting any more.

%d bloggers like this: