Below is a summary of my post-CPI tweets.
- 22 minutes until CPI. Not sure I am looking forward to this one. The little birds are all whispering that this is supposed to be high, and that concerns me.
- Not the economists: consensus forecast is for a reasonably high 0.24% on m/m core. But we drop off 0.22% from last December so the y/y won’t move much from 1.7% if that’s the print we get.
- Yes – there are lots of reasons this COULD be higher. Chief among them is the divergence between surveys of used car prices and the BLS cars index. Cars are 6.4% of CPI, so it matters. But PPI showed weakness in vehicles for another month. (I usually ignore PPI, though).
- ..it’s December, which means it’s crazy-seasonal-adjustment month. December is the only month of the year where you can confidently reject the hypothesis that there’s no seasonal (on headline CPI), as prices tend to fall. But there’s also a lot of volatility.
- Rents have come back to model, and home prices continue to rise, so decent chance that housing starts to contribute again here soon.
- What I fear is that some of the forecasts for a “surprise” higher are coming from the fact that the inflation markets have been rallying, so people are afraid “someone knows something.” Economists don’t ignore markets. But in this case I think it’s just year-end reassessment.
- …let’s face it, inflation bonds are cheap. About 50bps cheap at the 10-year point by my model. Commodities are cheap. And everything else is expensive. I don’t have to believe inflation is coming to swap out of stocks into commodities.
- Of note – inflation swaps have been rising in every major market recently. So there definitely is an undercurrent of inflation concern.
- Don’t fade the whispers! +0.3% on core. Actually 0.277%. But enough to put y/y up to 1.77%, rolling it to 1.8% rounded.
- Wow, 2 yr Tsy above 2% for the first time since September 2008!
- Last 12 CPIs. Try hard not to see an uptrend here. It’s an illusion caused by the low mid-year figures. But that said, this is highest in a while.
- Let’s see…Housing up slightly, Transportation up, no change in medical care (talking major subgroups here)…will be interesting to see where the wiggle is.
- Core services 2.6% vs 2.5% and core goods -0.7% vs -0.9% y/y. That’s the least goods deflation since last July. But it’s still deflation.
- Pulling in the micro data now. The BLS series is so rich. But while the sheet is calculating this is a good time to remind everyone that these figures are for DECEMBER so try hard not to get too excited. The breakdown will be more important to tell us if this is ‘real.’
- If you haven’t read Ben Inker’s piece in the latest GMO quarterly, arguing why inflation is a bigger risk for portfolios right now than recession, do so. Very good piece. “What happened to inflation? And What happens if it comes back?” https://www.gmo.com/docs/default-source/research-and-commentary/strategies/gmo-quarterly-letters/what-happened-to-inflation-.pdf?sfvrsn=5
- One more item of context before we dive deeper: Median CPI is at 2.3%. So we should be expecting something right around 0.2% per month if there’s no trend. The uptick from 1.7% to 1.8% is just catching up, mostly.
- OK on the breakdown. New and Used cars, 8% of core CPI, rose to -0.33% from -1.05%. As expected, and that’s a big part of the surprise.
- I say “surprise,” but it really oughtn’t be a surprise. Remember that Hurricane Harvey had a similar effect to Cash for Clunkers in terms of the number of cars removed from the road. The private car prices indices were showing this. BLS has a lot of catching up yet.
- Just lost power. Anyway. Wasn’t just used cars. Used cars went from -2.1% to -0.99% but new went from -1.08% to -0.53%
- Owners Equivalent Rent went to 3.175% from 3.124%, and primary rents from 3.675% to 3.689%. So housing back on track.
- Medical Care broadly went to 1.78% from 1.68%. Pharma went 1.87% to 2.37%. Other components pretty stable (in medical). Medicinal drugs (pharma) is about one fifth of medical care subindex.
- Wireless telephone services again steady. The jump will be fun when the plunge washes out. Right now it’s -10.19% y/y vs -10.24%.
- Wish I could post my chart of distributions of price changes. Left tail starting to move rightward a bit. Hopefully get power back soon. This is all on backup power to my pc. [Editors’ Note – I added it later, see below]
- Well, looks like power isn’t coming back on quickly. I will have to come back later with the median CPI estimate etc. Got most of the details out though.
- Bottom line is that the components we expected to start converging, did. Housing behaved. Medical care behaved. And so we moved towards the real middle of the distribution, around 2.3% or so presently.
- This shouldn’t be taken as an acceleration in inflation. This is just one (flawed) number converging with the better ones. Core inflation is going to head higher, but this isn’t convincing evidence that it is yet doing so.
- Having said that, in a couple of months the y/y comps start to get better so the inflation story will have much better OPTICS. And it’s optics these days, more than fundamentals, that drive markets. So don’t jump off the commodities or tips bandwagon. That trend will continue.
- Power’s back on! Of note is that Median CPI printed at 0.29%, the highest level since July 2008 (sound familiar? That was also true two months ago when it was 0.27%). So y/y up to 2.44% now.
- Yeah, I know I said don’t think of this was an uptrend. And it’s not; it’s an unwind of one-offs. But still, that’s gotta look scary.
- Better late than never. Here’s what I meant about the distribution moving right. Those two bars on the left were one bar before today. So you can see those components – largely cars and cell phones – are dragging down core relative to median.
- The rally in breakevens shouldn’t be terribly surprising – this chart shows it’s just keeping pace (and not even) with the turn back higher in median CPI.
- The market is NOT AT ALL ahead of itself in this sense.
This was certainly not the easiest time I have had with a CPI report, but that’s mostly because the power grid in this country is as brittle as glass. The story was actually not as much about screwy seasonal as I was concerned about. Actually, it was a fairly humdrum report in many ways, and that’s what is scary if you’re thinking we are in a “lowflation” period. The chart of Median CPI is interesting. Core inflation had risen mostly because car prices are starting to catch up with private measures of car prices – what remains in the gap between the red line and the blue line in the “Manheim” chart would add about 0.5% to core CPI – and housing stopped decelerating. But then Median CPI, which doesn’t care about the New and Used car prices since those are outliers, rose at the highest rate (m/m) in nearly a decade, and the Median-Core spread actually widened slightly this month. That means more core acceleration is ahead.
I mentioned that in a few months the year-ago comparisons will start getting easier. This month, we got 0.28% from core CPI versus 0.22% last year. But in Jan 2017, core CPI was +0.31%. That will be a hard comp to beat. But after that, Feb 2017 was +0.21%, March was -0.12%, April was +0.07%, May was +0.06%, June was +0.12%, and July was +0.11%. At the time, we mused “is the natural run rate for core really 0.5%/annum?” which was what those five months were averaging. That seemed very unlikely. Median CPI told us that wasn’t the case. Now, if core CPI merely averages a monthly 0.17% print from now until July, the y/y figure will be up at 2.20%. And if it’s 0.2% per month, in July we will be sitting at 2.42%.
I don’t think you want to fade those optics, even if you think we’re only going to get 0.15%. Perhaps the next month or two, because of the more-difficult comps, will take some wind out of the sails of the inflation bulls and offer better entry points. But the direction of travel looks fairly clear for the next six months or so. And that also means that the direction of travel for monetary policy is also likely set, to be at least as aggressive as the market is pricing. And, perhaps, the direction of travel for equity prices isn’t quite as clear as it currently seems.
And it bears repeating that this is going to be the case even if inflation is not actually in an uptrend, but just maintaining its current run rate around 0.2% per month (commensurate with median CPI at 2.4%/yr). If inflation is in fact turning higher – and there are some signs of that, though not as widespread as everyone seems to suddenly think – then it could be a lot uglier in 2018. As I said again above: don’t jump off the commodities or TIPS bandwagon yet. But…you might want to trim some of that nominal bond exposure!Subscribe to NFTRH Premium for an in-depth weekly market report, interim updates and NFTRH+ chart and trade ideas; or the free eLetter for an introduction to our work. You can also keep up to date with plenty of actionable public content at NFTRH.com. Or follow via Twitter @BiiwiiNFTRH, StockTwits or RSS. Also check out the quality market writers at Biiwii.com.