Every conversation about AI eventually runs into the same wall: power.
You’ve heard the version everyone repeats. Data centers need electricity. The grid can’t keep up. Utilities are spending record amounts. Transformers have a two-and-a-half-year lead time. So investors pile into the obvious names, the transformer makers, the switchgear companies, the electrical contractors, and pay thirty or forty times earnings for the privilege.
Here’s the part that gets skipped.
Before any of that electricity reaches a server, it has to travel. Down transmission lines. Through substations. Across thousands of miles of conductor strung between poles and towers. And every single foot of that wire has to be gripped, spliced, anchored, protected from vibration, and held in place by physical hardware.
That hardware is not glamorous. It’s rods, clamps, dead-ends, splices, and closures. Small steel and aluminum parts. Nobody writes think pieces about them. But federal data shows more than 7,000 circuit miles of new and upgraded transmission lines entered service in a single recent year, and the vast majority of it wasn’t the marquee ultra-high-voltage stuff. It was ordinary line miles. Lots and lots of them.
Somebody makes the hardware for all of it.
That somebody is a company founded in 1947, in Cleveland, by an engineer who solved a problem nobody else had bothered to solve. Its US sales just grew 32 percent in a single quarter. Its earnings per share nearly doubled. And the stock recorded its largest one-day move on record, up roughly 30 percent, the morning after it reported.
Most investors have never heard the name. There’s a structural reason for that, and it is the entire opportunity.
But here’s where it gets strange.
This company doesn’t hold an earnings call. It doesn’t issue guidance. It doesn’t tell you its quarterly backlog. As far as I can tell, exactly one brokerage firm on Earth publishes a rating on it, and that firm has changed its mind three times in five months.
Which means there is no consensus estimate. No revision cycle. No sell-side chorus slowly repricing the story over six months.
There is only the press release, dropped into the void once a quarter.
Below the paywall: the name, the full quarter, the grid math almost nobody is doing, the one dividend decision that told you the family running this business changed its mind about the future, and the four reasons this could go very wrong. This one has a bear case worth reading first.
