We're finally starting to see hints of relief when it comes to inflation. Prices at the pump are starting to come down, monthly core CPI momentum has slowed, used cars were down around 2% YoY, and food inflation is starting to moderate. On the other hand, there's America's massive explosion in artificial-intelligence infrastructure - which is beginning to push prices up on everything from electricity to smartphones.
On Thursday Apple announced a 15-25% price hike on Mac computers and iPads, after CEO Tim Cook told the Wall Street Journal that the jump in costs was unlike anything he had seen "in any area in over 40 years." An Apple spokesperson placed the blame on the "rapid expansion of AI data centers, which has created an extraordinary surge in demand for memory and storage," causing component prices to surge.
Elon Musk agrees...
As the Wall Street Journal notes;
The money pouring into the AI arms race is unprecedented. Analysts peg capital spending at five of the so-called hyperscalers—Alphabet, Amazon, Meta Platforms, Microsoft and Oracle—at $741 billion this year, according to FactSet, up nearly 75% from last year.
Where is all that money going? While much of the conversation is focused on what AI can do, the build-out itself is strikingly physical, said Columbia University economist Stijn Van Nieuwerburgh. -WSJ
AI data centers require specific, sophisticated equipment to ensure cool, stable operation - as well as electric and fiber-optic cables and backup generators in order to keep them running 24-7. According to the report, Van Nieuwerburgh estimates that the AI buildout could cost somewhere in the range of $8 trillion over the next six years. As such, the demand for components shared throughout the economy (memory, for example), the effects are now trickling down to consumer electronics - like iPads. Other companies such as Nintendo, Microsoft and Sony have all raised prices on devices.
According to the Labor Department, consumer prices for computer software and accessories were up around 15% from a year earlier in May, while the Department's measure of wholesale electronic components and accessories shot up 27% from a year earlier last month.
When it comes to electricity - the price began to rapidly increase during covid - and it's now slingshotting even higher. Note the rate of change in the lower panel.
According to Goldman, data centers will account for nearly half of US growth in power demand through 2030 - and see consumer electricity prices rising around 6% annually in 2026 and 2027.
The Journal also notes that while tariffs and oil were one-time economic shocks, the AI shock to demand could persist for years.
That dynamic is reflected in the rally in the shares of chip stocks, which have moved sharply higher on investor expectations of sharply higher demand. Even with a sharp selloff this week, the PHLX Semiconductor Index is up about 150% over the past year.
Of course, more than just chips go into data centers. And like chips, a lot of the other things that go into building and running a data center are used widely across the economy. That could raise costs for a variety of businesses, which may then try to recoup those costs by charging consumers higher prices.
In some instances, the AI build-out could also add to labor costs. Wages for workers who are in demand from data-center construction have been picking up: Average hourly earnings for electrical and wiring-installation contractors were up 6.5% in April from a year earlier, which compared with 3.6% for all private-sector workers. -WSJ
Still, economics aren't predicting an AI-fueled inflation surge like we saw during Covid.
In November, now-Fed Chairman Kevin Warsh wrote in a WSJ op-ed that "AI will be a significant disinflationary force, increasing productivity and bolstering American competitiveness," arguing "productivity improvements should drive significant increases in real take-home wages. A 1-percentage-point increase in annual productivity growth would double standards of living within a single generation."
Yet, UBS economists think that the delta between the current building frenzy and AI lowering prices will be at least a couple of years.
According to a Monday survey by the National Association for Business Economics, 81% of those polled said the AI build-out will add to inflation over the next year.
"In the first phase of any major technological revolution, you tend to have a strain on limited resources, and that tends to put upward pressure on prices," EY-Parthenon chief economist Gregory Daco - president of NABE - told The Journal.
TL;DR - the AI build-out may keep inflation broadly elevated, and at some point it may all be worth it in the form of disinflationary productivity. Then again, who's going to buy anything when tens of millions are without jobs that are now done by AI?