IES Holdings: The Market Is Pricing a Data-Center Slowdown Its Backlog Doesn't Show

IES Holdings hit an all-time high of $816.51 on August 12. This week the shares have changed hands near $310.

Before the confusion: IES split its stock 2-for-1, and split-adjusted trading began on August 24. Even accounting for the split, the stock is down about 24% from the August 12 record — roughly 17% over the last month and close to 10% in the last five sessions. This is not one bad day. It is a repricing, and it has happened in step with a marketwide rethink of the AI data-center trade.

IES Holdings, a Houston contractor that designs and installs electrical, communications, and technology systems, rode that trade from a 52-week low near $167 to just over $408 at the August peak (both split-adjusted) — more than doubled in roughly a year. The selloff's fuel is a slowdown story that, on the surface, is about politics. At least 12 states have introduced data-center moratorium bills. Governors in Pennsylvania and Texas who once welcomed the buildout are calling for stricter rules. Counts of delayed or blocked projects run from $64 billion to $130 billion. On August 24, Jim Cramer told CNBC viewers the trade was under political attack and that "the unbridled buildout is most likely over." Two days later, asked about IES directly, he said something quite different: IES is "a very interesting data communications company that's been brought down recently with it, and I don't think that makes sense. I think you've got a good one."

A TV host's blessing is not an investment framework, and Cramer has been long this name since it was a fraction of today's price — "a buy" in November, "going higher because it is data center" last September — so his enthusiasm is not fresh information. But the divergence he keeps pointing at is real: IES just reported a quarter that obliterated expectations, and four weeks later the stock is down a quarter from its peak.

The quarter the slowdown story has to explain

Revenue of $1,242.7 million, up 40% year over year. Operating income of $178.5 million, up 60%. Net income attributable to IES of $153.0 million, up 98%. Adjusted diluted EPS of $6.70 — pre-split dollars, so $3.35 per current share — versus consensus of roughly $4.80. The stock jumped about 30% the session after the report. Over nine months, revenue is up 25% to $3.09 billion and net income up 70%, and the company raised its fiscal-2026 capital-spending plan.

Now the number that separates the slowdown story from the slowdown itself: backlog of about $4.5 billion at June 30, up 91% since the end of fiscal 2025. For a contractor, backlog is contracted work not yet built — revenue already in hand. $4.5 billion is roughly three and a half quarters of IES's current revenue. If the slowdown were real at this company, the first visible sign would be an order book that stopped growing. It hasn't. In its most data-center-heavy segments, Commercial & Industrial revenue rose 109% and segment operating income went from $12.9 million to $54.2 million; Communications revenue rose 51%, which the company credits to the data-center market.

The market's case isn't imaginary

The stock was expensive, and expensive by design. Even after this month's fall it sits near 27 times trailing earnings, roughly 3.1 times sales, and about 22 times trailing EBITDA — versus a plain-vanilla specialty contractor like MYR Group at about 1.1 times sales. IES earns that premium with margins and growth, but a premium multiple meeting a narrative that can dent growth is the exact recipe for a fast air pocket. The split added mechanics: cosmetic for value, but the first post-split session traded at about a third of normal volume and fell 5% — the kind of thin, low-conviction tape that lets a trend run.

And there are two genuine bear facts that must be answered, not waved away.

Residential is deteriorating, not just soft. Revenue fell 6% and segment operating income was cut nearly in half, to $16.3 million, on housing softness and a limited ability to pass material-cost increases through. That is the cyclical floor of this company, and it is moving the wrong way.

The DBM Global deal changes the balance sheet. On August 10, IES agreed to buy the structural-steel fabricator for a headline price of about $650 million — roughly $685 million including tax-election costs — funded with about $545 million in cash and $140 million in stock. DBM brings roughly $1.3 billion of revenue and 3,400 employees into a new fifth operating segment, and the cash side will be financed in part through an expanded credit facility. IES entered the quarter debt-free with $77 million of cash and $311 million of marketable securities; it may finish the year with a genuine debt load, integrating a lower-margin business just as the sector's narrative turns. The market's initial response — shares fell when the deal was announced — was not naive. And the executive chairman's expectation that cash flow repays the debt quickly is a reasonable bet for a cash-heavy builder, not a certainty.

Where the two sides actually disagree

The bull case is that the market is pricing a slowdown that does not appear in IES's own order book. The bear case is that the book is a rearview mirror on a permitting cycle about to bite. Both cannot be right, and the calendar will sort them: whether backlog keeps growing into the fiscal Q4 report expected in December, whether bookings flatten to about a book-to-bill of one, and whether Residential keeps falling while the DBM financing is finalized. Those are the checks that decide if the market is punishing a good company or correctly grading a peak.

On price, there is no case for catching the knife while it drops 10% in five sessions and swings 6% a day; the better risk/reward is the stabilization, not the decline. And there is equally no case for letting a sector tide write your fundamental opinion: if the order book keeps filling, selling a genuine compounder because the whole complex got repriced is letting the market's worst guess do your work.

Here is the honest summary, without the rating theater. The price is disappointing and the business, so far, is not — and the debate over this stock is a refusal to keep those two claims separate. If you believe the slowdown is real, find it in the order book; as of June 30 it isn't there. If you believe the growth persists, remember the price still demands you be right: at 27 times trailing earnings, the market has not baked in doom, it has simply stopped paying up. Even the aggregate signal from AInvest — a Hold — is not a buy-the-dip stampede. Both burdens are real, and the December quarter is where one of them gets discharged.