US consumer prices rose in January by more than expected compared to the same month last year. At 6.4% and 5.6%, both the headline and core gauges topped estimates.
But increases from the prior month were generally consistent with expectations. The headline index rose 0.5% and core 0.4% from December, according to heavily scrutinized data released on Tuesday.
The report included updated category weightings and methodological changes, and came on the heels of upward revisions for figures covering the final months of 2022. In other words: The data was quite difficult to parse.
The 0.4% advance on the core gauge was the second consecutive, and I’d gently note that such readings aren’t necessarily consistent with a rapid return to the Fed’s definition of price stability.
The half-point MoM increase on the all-items index was the hottest since October.
Not surprisingly, shelter was “by far the largest contributor,” as the BLS put it, accounting for almost half of the monthly gain.
You could look at that one of two ways. It’s bad news that measured inflation for roofs (if you will) is still running very hot. On the other hand, it’s widely understood that these prints are backward-looking and destined to fall in the months ahead.
The MoM gain on the shelter gauge was 0.7%, down from December, but still near the fastest monthly pace in four decades.
The YoY rate, at 7.9%, was the hottest since June of 1982 (but who’s counting, right?)
Context is important. As discussed at some length in “Disinflation… If You Can Keep It”+, asking rents are now rising at a slower pace than the overall cost of housing. That’s very notable. In March of last year, asking-rent inflation was running almost 17.5%. That figure is around 5% now.
Jerome Powell and other Fed officials have repeatedly referenced “pipeline disinflation” in housing, which is just a needlessly academic way of saying that to the extent CPI is a reflection of shelter costs, we’re looking at last year’s inflation.
Moving along, used vehicle prices fell again in January. The 1.9% MoM drop was the seventh consecutive, and brought the YoY pace of used vehicle deflation to nearly 12%.
This too needs context. The Manheim index rose last month, a development described as “not typical.” “The higher conversion rate indicated that [January] saw sellers with more pricing power than what is typically seen for this time of year,” the report said.
That stoked fears that a key disinflationary tailwind was dissipating+. The reported decline in the CPI used vehicle series thus came as something of a relief.
As expected, gas prices rose from December, but at just 1.5%, the YoY increase looked tame, and base effects should be a factor going forward. That, even as Russia’s decision to cut oil output and the never-ending “Chinese re-opening demand” story continue to inform many a bull case for crude.
Natural gas prices posted a sharp jump from December. That series is (obviously) quite volatile. Electricity notched another monthly gain, and YoY, the index was up almost 12%.
Another meaningful MoM increase in food at home prices left the 12-month rate at 11.3%. So, Americans continue to experience double-digit inflation for electricity and groceries simultaneously. That’s very rare. Or at least it was rare. Now, apparently, it’s the norm.
To be fair, the YoY prints on both the food at home and electricity gauges topped out in August at 13.5% and 15.8%, respectively.
Elsewhere, gauges for new vehicles, transportation services and apparel all rose. The 0.8% increase on the apparel index was the biggest monthly jump since December of 2021.
All in all, the report certainly wasn’t encouraging, but at the same time, there was rampant speculation that the figures would come in markedly hotter than expected, thereby amplifying the hawkish read-through for monetary policy from the scorching US jobs report. It wasn’t immediately obvious that the figures lived up to that billing, although if you strip out the big drop in the medical care index, services inflation rose 0.8% last month.
Certainly, the numbers reinforced the case for two additional Fed hikes, and maybe even three more — or at least that’s what policymakers will say. But I don’t think the figures matched the level of market angst headed into Tuesday.
I just noticed that housing prices in cities like Mesa Arizona are dropping as the water supply drops. I don’t see any efforts being initiated to build new sustainable supplies. People are starting to see the writing on the wall. They want to sell first while there may be some demand… even at a loss. From Utah to New Mexico and California, areas running out of water will start seeing a migration out. This will keep housing prices high in areas that are not facing water shortages. Demand will go up. Similar to areas that are facing rising sea levels and increased flooding, the migration is under way.
Residential water usage isn’t really the issue in those states. The vast majority of the water in California and Arizona is actually used for agriculture. It’s roughly 80% and 74%, respectively. If push comes to shove, agricultural water usage has to be reduced. That’s not to say specific locales won’t have water issues or that there aren’t implications for food costs or future development, but these states are certainly investing in ways to make better use of their water. In fact, per capita water usage in California is down significantly over the last 50 years. One study I saw estimated that total water usage has gone up only 20% in the last 50 years despite the population more than doubling in that time.
I can see why people would leave Arizona though since it’s hot as hell and only going to get hotter for several months out of the year.
“Through the roof you can’t afford”… Haha. An instant classic in my book!
This question is off topic, but do you have any plans/ideas to comment on the current stir around AI and ChatGPT at any point in the near future?
Not from a short term speculative point of view, but the structural/long term effects of that technology and how it might tie into deflationary trends, UBI, National Security etc.
If I read that correctly, these new header illustrations are all AI generated, no? I wonder what experience you have made using it.
Cheers
Looking at Case-Shiller house price index vs monthly CPI shelter
https://fred.stlouisfed.org/graph/?g=104Ko
In the Great Recession, house price peaked June 2006 and monthly shelter CPI peaked in late 2006. Looking at 2000 is unhelpful, as house price didn’t peak. In 1990 house price and monthly shelter CPI peaked about the same time. Pre-1983, the shelter CPI methodology was completely different. https://www.whitehouse.gov/cea/written-materials/2021/09/09/housing-prices-and-inflation/
House price peaked June 2022, if the 2006 relationship holds monthly shelter CPI should have peaked in late 2022, and if the 1990 relationship holds we’d have seen monthly shelter CPI peak last summer.
It is not implausible for the lead/lag relationship between house price and monthly shelter CPI to be different in 2023 from earlier peaks, just as 2006 was itself different from earlier.
Shelter CPI is mostly Owners Equivalent Rent CPI. OER is a survey of homeowners’ estimates of how much their house would rent for.
How much attention does the average homeowner pay to what other houses are renting for? Maybe about as much as the average car owner pays to what other cars are renting for. I have only the vaguest idea of the range of possible rents for houses in my neighborhood; it is wide ($3000 to $5000), formed from only a handful of listings (whatever I’ve happened to notice over the last decade), and infrequently updated (since I don’t care).
Perhaps responses to this survey are more guided by how much they think their house is worth or by impressions of overall inflation. In a period when existing housing sales volume and existing house on-market inventory are both very low, sellers are “in denial”, while overall inflation is high, perhaps OER can stay detached from Case-Schiller house prices for longer than previously.
Following up, perhaps other homeowners pay more attention to their local house market than I do. In which case, they might track not just listing and selling prices, but also mortgage rates. In a period when existing house prices decline but mortgage rates rise, perhaps their responses are guided by mortgage payments. Given how those have risen, perhaps OER can also stay detached for longer.
More broadly, why should the shelter price of houses track the selling price of houses? Should it instead track the periodic cost to the homeowner of occupying the house, e.g. mortgage payment, property tax, maintenance, excluding utilities and other occupancy costs captured in other CPI components?