Carrier's data-center cooling story masks a 1% organic decline and a 44x multiple

---

On April 30, 2026, Carrier Global Corporation reported its first quarter and the market chose, as it almost always does, the most flattering sentence in the release. Global data-center orders were up 500% year over year. The backlog, management said, already covered the entire $1.5 billion data-center sales target for 2026. The stock rose roughly 9% in the session. To read the coverage that followed, you would think Carrier had quietly stopped being a maker of air conditioners and furnaces and become an artificial-intelligence infrastructure play — a thermal-management supplier riding the same hyperscaler capital wave that has minted fortunes at Vertiv and Nvidia. That is the story embedded in the multiple. As of early June 2026, Carrier traded at a trailing price-to-earnings ratio near 44, against a ten-year median closer to 18. The re-rating is not a rumor. It is sitting in the price.

This piece is about the distance between that sentence and the rest of the income statement. Because the same quarter that produced the 500% headline also produced organic revenue down about 1%, a GAAP operating margin that fell from 12.1% a year earlier to 4.8%, net income that dropped from $412 million to $238 million, adjusted EPS down 12% year over year, and free cash flow of negative $15 million. The residential heating and cooling business that remains the company's economic core saw CSA Residential sales fall roughly 12% as distributors worked down dealer inventory. The forensic question is not whether the data-center orders are real — they appear to be. It is whether a single fast-growing sliver, bolted onto a slow, cyclical, housing-tethered base, justifies pricing the entire enterprise as a secular growth compounder. The order book is a demonstration. The thesis requires deployment, at margin, for years, while the rest of the business does not deteriorate faster than the new engine can grow. That is a great deal of perfection to pay 44 times earnings for.

The denominator illusion: 500% of a small number is still a small number

Begin with the headline itself, because it is constructed to overwhelm scrutiny. A 500% increase in data-center orders is a genuinely striking figure. But a percentage is a fraction, and a fraction has a denominator. Carrier's data-center revenue crossed roughly $1 billion in 2025 and is guided to about $1.5 billion in 2026. Against a company whose full-year 2025 sales were $21.7 billion and whose 2026 revenue guide rounds to slightly above $22 billion, $1.5 billion of data-center revenue is on the order of 7% of the enterprise. The orders growing at 500% are orders into a base that, a year or two ago, was a rounding error. When you grow from a small number, enormous percentages are arithmetically cheap. The same is true of the prior quarter: management touted North American data-center orders up 400% in Q4. Four hundred percent, five hundred percent — these are the percentages of a denominator just large enough to matter and just small enough to flatter.

This is not to dismiss the segment. A business going from $1 billion to $1.5 billion in a year is adding real, high-quality revenue, and commercial HVAC — the segment that houses high-efficiency chillers and data-center cooling — is plausibly Carrier's best growth and margin story. But the re-rating prices the whole company as though the data-center denominator were the company. It is not. Roughly 93 cents of every revenue dollar still comes from everything else: residential and light-commercial heating and cooling, the Viessmann European climate business, refrigeration. And that 93% is the part of the franchise tied to housing cycles, European energy policy, Chinese consumer demand, and the replacement timing of furnaces and heat pumps — none of which is growing at 500%, and some of which is shrinking.

The organic stall the headline buries

Strip the segment theater away and look at the consolidated organic number, because it is the single most honest line in the release. Reported revenue rose 2.4% to $5.341 billion. But organic revenue — the figure that removes the contributions of acquisitions, divestitures, and currency, and tells you what the underlying business actually did — declined by about 1% year over year. The reported growth came from elsewhere. Pricing contributed; the math of the portfolio reshuffle contributed. The actual volume-and-mix engine of the existing business went backward.

That distinction is the whole forensic case in miniature. A company can report "growth" for years while its underlying operations stall, provided it keeps buying revenue and reshaping the denominator faster than the core erodes. Carrier has spent the last two years doing exactly that on an epic scale: it paid roughly $14.2 billion for Viessmann Climate Solutions and generated more than $10 billion in proceeds divesting Fire & Security, the commercial refrigeration business, and other units. When a company simultaneously adds a $14 billion business and sells off $10 billion of itself, the year-over-year revenue line becomes nearly uninterpretable without the organic figure — and the organic figure says the core contracted. The 500% data-center headline is doing a great deal of work to keep your eyes off a minus sign.

Adjusted versus GAAP: the $0.28 the press release does not lead with

Now the quality-of-earnings question. Carrier reported adjusted diluted EPS of $0.57, beating the roughly $0.51 consensus and driving the post-print rally. The GAAP diluted EPS from continuing operations was $0.28 — barely half. Net income fell to $238 million from $412 million in the prior-year quarter. And the GAAP operating margin, the cleanest measure of whether the business made money turning revenue into profit, collapsed to 4.8% from 12.1% a year earlier.

A 730-basis-point drop in GAAP operating margin in a single year is not a rounding artifact. Some of it reflects restructuring — Carrier announced roughly 3,000 job cuts — and portfolio costs that the adjusted figure strips out as non-representative. There is a defensible logic to excluding one-time separation and integration charges from a "clean" run-rate. But the habit of anchoring the entire investment narrative to a $0.57 number that is double the $0.28 the company actually earned under generally accepted accounting principles deserves the same scrutiny it would get at any company carrying this much integration baggage. The adjustments are not free. Restructuring is real cash. Integration of a $14 billion acquisition is real cash and real management attention. The adjusted EBITDA margin of 16.2% on $866 million looks like a healthy industrial. The GAAP operating margin of 4.8% looks like a business absorbing the cost of having been torn apart and stitched back together. Both are true. Only one of them is the number the multiple is built on, and it is not the GAAP one.

Cash does not lie: free cash flow went negative

The cleanest test of whether a "beat" is real is the cash. In Q1 2026, Carrier generated $79 million of operating cash flow and free cash flow of negative $15 million. A company being priced as a secular compounder at 44 times earnings produced, in the quarter that triggered its data-center re-rating, less than nothing in free cash.

Seasonality is a fair partial defense — HVAC is a working-capital-heavy business and the first quarter is typically its weakest for cash as it builds inventory ahead of the cooling season. Carrier will, in all likelihood, convert cash over the full year. But the direction of travel matters, and the year-over-year comparison is unflattering: this represents a meaningful decline from the prior-year quarter. A business that is both stalling organically and burning first-quarter cash, while carrying the integration cost of a $14 billion acquisition, is a business where the gap between the adjusted story and the cash reality is widening, not narrowing. The 500% headline does not generate a single dollar of free cash flow. The furnaces and chillers and heat pumps do, and in the quarter the market chose to celebrate, they net produced none.

Bought growth, sold core: the portfolio shuffle as a magic trick

Step back and look at what the last two years of dealmaking actually accomplished. Carrier paid roughly $14.2 billion for Viessmann, a German maker of heat pumps and climate solutions, to lever itself into European electrification and the energy-transition narrative. It then raised more than $10 billion selling Fire & Security, commercial refrigeration, and other businesses. The official framing is "transformation": a sprawling conglomerate refocusing into a pure-play "intelligent climate and energy solutions" company.

The forensic framing is less flattering. The company bought a growth story and sold cash-generating diversification, financing a re-rating with the balance sheet and the income statement's comparability. Viessmann arrived with a hefty multiple and was expected to deliver only about $100 million in incremental revenue synergies in 2025 — a modest figure against a $14.2 billion price. Carrier has since agreed to sell its Riello business for about $430 million, an acknowledgment that not everything acquired or owned fits the new story. When the underlying business is stalling, portfolio reshaping is a powerful tool for keeping the reported numbers moving in the right direction: it changes the denominator, it changes the segment mix, it gives the market new narratives to price. But it does not change the fundamental cyclicality of selling heating and cooling equipment to homeowners, builders, and European consumers. It just makes that cyclicality harder to see for a few reporting periods.

Cyclical priced as secular: the housing tether nobody re-rated away

Here is the core of the asymmetry. The bull case prices Carrier on the data-center and electrification secular themes. The actual business is overwhelmingly a bet on housing, building, and replacement cycles that have not been repealed.

The evidence is in Carrier's own Q1 results. CSA — the climate-solutions-Americas residential and light-commercial business — saw organic sales decline about 3%, with CSA Residential sales down roughly 12% as distributors cut purchases to rebalance dealer inventories. That is the textbook signature of a housing-cyclical business: when home sales slow, when builders pull back, when consumers defer the discretionary replacement of a working furnace, the distributors who stock Carrier's equipment stop ordering, and the manufacturer feels it immediately. Layer on persistent weakness in China Residential and Light Commercial, and a meaningful chunk of the enterprise is contracting for reasons that have nothing to do with artificial intelligence and everything to do with interest rates, affordability, and the Chinese consumer.

A company trading at 44 times earnings is priced as if its earnings stream were durable, growing, and insulated from the cycle. A company whose largest segment just printed a 12% residential decline is priced, by the cycle, as exactly what it is: an equipment maker at the mercy of demand it does not control. You cannot have it both ways. Either the housing exposure is real — in which case the multiple is a cyclical peak dressed as a growth re-rating — or the data-center business is so transformative it swamps the housing drag, which the 1% organic decline flatly contradicts.

Demonstration versus deployment: backlog is a promise, not a margin

The most seductive line in the release is that the data-center backlog already covers the full-year sales target. Treat that with the skepticism backlog deserves. A backlog is a record of orders booked, not revenue earned, not cash collected, and certainly not margin realized. It is a demonstration of demand. It is not yet deployment, and the distance between the two is where industrial theses go to die.

Data-center cooling is a fiercely competitive arena. Vertiv, Trane, Johnson Controls, and a field of specialists are all chasing the same hyperscaler thermal budgets. Winning orders in a land-grab phase, when hyperscalers are placing equipment orders ahead of construction, tells you a vendor is competitive on price and availability. It does not tell you the work will convert to the margins the multiple assumes, or that the order pace persists once the current build-out wave normalizes. Hyperscaler capital expenditure is itself cyclical and concentrated among a handful of buyers; an order book that swung 500% up can swing the other way if a few customers pause. The bull narrative quietly assumes that today's demonstration of order strength becomes years of high-margin deployment. That is a forecast, not a fact, and at 44 times earnings the market has already paid for the optimistic version.

Customer concentration and the policy crutch

Two further dependencies sit under the secular story. First, concentration: the data-center demand Carrier is celebrating flows from a small number of hyperscalers whose capital plans are set quarter to quarter and can be throttled with a single budget decision. A growth engine whose order book depends on perhaps a handful of buyers is structurally more fragile than the diversified furnace-and-chiller base it is meant to replace in the narrative.

Second, policy. The Viessmann electrification thesis — heat pumps displacing gas boilers across Europe — leans heavily on subsidy regimes and decarbonization mandates that are politically contingent and have already proven volatile. European heat-pump demand softened sharply when subsidy expectations wobbled. A growth story underwritten by government incentive schedules is a story with a regulator's hand on the throttle. None of this makes the businesses bad. It makes them cyclical and policy-dependent — the opposite of the durable, self-propelling secular compounder a 44-times multiple implies.

What the bulls genuinely get right

The bull case here is not foolish, and a fair forensic account must say so plainly. Several things in Carrier's story are genuinely strong, and the skeptic who ignores them will be wrong.

First, the data-center order strength appears real and large, not a one-quarter accounting flatter. A backlog that covers the full-year $1.5 billion target is a tangible demand signal, and commercial HVAC — high-efficiency chillers and data-center cooling — is plausibly entering a sixth consecutive year of double-digit growth. Thermal management is a genuine bottleneck in AI infrastructure, and Carrier is a credible, scaled supplier into it. That is not a fantasy; it is a real secular tailwind that a real franchise is positioned to capture.

Second, the installed base and services annuity are underappreciated by the pure housing-cycle bears. Carrier sells equipment that runs for fifteen to twenty years and generates recurring service, parts, and aftermarket revenue with margins and stability far better than new-equipment sales. That installed-base annuity dampens cyclicality and is a legitimate source of quality the GAAP-margin snapshot in a transition quarter understates.

Third, the portfolio transformation, whatever its multiple cost, did produce a more focused company with a cleaner climate-and-energy identity, real European scale through Viessmann, and over $10 billion of proceeds to deploy into buybacks and debt reduction. Management is cutting costs — the roughly 3,000 job reductions — to defend margin through the residential trough. And the residential weakness driving the organic decline is, in part, a destocking phenomenon: distributors rebalancing inventory, not end-demand collapsing. Destocking reverses. If housing stabilizes and the data-center ramp holds, the 2026 trough could mark the bottom of the organic line rather than the start of a slide. The bull is betting on exactly that inflection, and it is a defensible bet. The disagreement is about price, not about whether the business is good.

The kicker

Strip the release to its arithmetic and a clear picture emerges. Carrier sold $5.341 billion of equipment in the first quarter, earned $0.28 a share under GAAP, watched its operating margin fall by more than half, burned $15 million of free cash, saw its residential core shrink 12% and its consolidated organic revenue go negative — and the market lifted the stock 9% because data-center orders, on a base near 7% of the company, grew 500%. The forensic case is not that the data-center business is fake. It is that a fast-growing sliver does not transmute a slow, cyclical, housing-and-policy-tethered equipment maker into a secular compounder worth 44 times trailing earnings, roughly two and a half times its own decade-long median. The order book is a demonstration of demand; the multiple is a bet that demonstration becomes high-margin deployment for years while the 93% of the company that pays the bills does not deteriorate underneath it. That is a great deal of perfection priced in for a business whose own organic line just printed a minus sign.

Carrier bought a growth narrative for $14.2 billion, sold $10 billion of its old self to pay for it, and now trades as if the cycle that built it had been repealed — but the residential furnaces still answer to housing, the heat pumps still answer to Europe's subsidy schedule, and the only line in the quarter that grew 500% is a denominator small enough that the rest of the company can still drag the whole thing down before deployment ever catches the demonstration.