This article is an on-site version of our Unhedged newsletter. Sign up here to get the newsletter sent straight to your inbox every weekday
Welcome back. After two weeks of writing Unhedged, I’m exhausted but exhilarated. Hope you’ve enjoyed the ride. Email me (robert.armstrong) and have a relaxing long weekend.
The inflation suspense is killing me
Last week I mentioned inflation worries, only to wave them off. The last big US inflation report looked like you would expect it to look in an economic reopening. Only sectors that had been crushed in the pandemic showed dramatic inflation. So the Federal Reserve still seemed unlikely to tighten policy, or even talk about tightening it, anytime soon. So: everybody be cool!
This week being cool got just a bit harder. On Thursday we received a strong jobless claims report (406,000 new claims, down from 444,000 last week and spitting difference from the pre-pandemic level of about 220,000). And on Wednesday the Fed’s vice-chair Randy Quarles said that:
“If my expectations about economic growth, employment and inflation over the coming months are borne out . . . and especially if they come in strong . . . it will become important for the [Federal Open Market Committee] to begin discussing our plans to adjust the pace of asset purchases at upcoming meetings.”
That’s what we call talking about talking about tightening.
Still, we shouldn’t move to a war footing. The inflation data is still binary, with stuff that is generally volatile, commodity-based or reopening-related spiking, while the stable consumer-y stuff is still in the range where the Fed is comfortable.
Here is a nice chart from Strategas showing a measure of consumer price inflation compared with an index of inflation in prices paid by manufacturers for inputs:
One line is quite scary, the other not very (since Don Rismiller of Strategas is more nervous than I am, I should note what he says: “The increases in inflation and surveys on inflation expectations are becoming difficult to ignore.”)
On the same theme, here’s a picture of the Atlanta Fed’s sticky (stuff like insurance, booze, cell phone bills) and flexible (gasoline, milk, hotel rooms) consumer price inflation measures. The orange line is the sticky one:
Again, the volatile stuff doesn’t seem to be having a strong gravitational effect on the stable stuff. If that changes, look out, but to my eye it is not changing yet.
Where does that leave us? Hanging. Here is the S&P 500 plotted against expectations for inflation in the five-year period starting in five years, that is, long-term inflation expectations:
At the risk of over-interpreting the chart, the market ran up despite rising inflation expectations all through the recovery — until late April. Then fear of inflation set in. The market has moved sideways since, as have inflation expectations. The two have fought to a tie. At some point the tiebreaker will be played.
Housing again, now with more data
A few days ago I wrote about how the recent acceleration in house prices might be indicative of a benign imbalance of supply and demand that has been exacerbated by the pandemic, or it might be evidence that speculative excess is creeping into housing. If it’s the latter, that’s bad. We don’t want multiple, interconnected speculative euphorias going on at once.
A standard proxy for speculative activity in housing markets is the number of homes purchased not for owner occupancy, that is, purchased to flip or rent. I didn’t have timely data on that the other day, but as it turns out the Mortgage Bankers Association tracks it. Here is that data, going as far back as the MBA has it, through to February of this year:
Something really changed at the end of April last year in the purchase market. It might just be people buying vacation homes to get out of the city! But 14 per cent is a big slice of the market. For comparison, here is some longer-term data, collected by the Princeton economist Wei Xiong and his colleagues:
At the height of the housing bubble, the proportion of all house purchases that were not primary homes hit 15 per cent. If these two data sets are equivalent, this metric is now at bubble levels. But I’m not sure they are equivalent. The MBA data is survey-based; the Xiong data comes from application disclosures collected by the government.
Does that mean we are heading into a second housing bubble? No. For one, buyers are much less leveraged today than in 2005. And, like all economic data right now, we could be looking at a pandemic anomaly in the MBA data, one that could abate. Mortgage applications and pending home sales have recently started to fall. Home prices and non-occupant purchases will, hopefully, fall with them.
Two good reads: something smells
In this weekend’s FT, I write about perfume. If the piece is as interesting for you to read as it was for me to write, check out this excellent article about smells that ran in The New Yorker earlier this year. Once you start thinking about this stuff, it’s addictive.