Carrier Global

Phase A — Understand the business

Company Overview

Carrier Global is a global leader in intelligent climate and energy solutions — in plain terms, the company that makes air conditioners, heat pumps, heating systems, building controls, and transport refrigeration, plus the aftermarket parts and service that follow them through their lifecycle. FY2025 net sales were $21,747M, down 3% YoY, with operating profit of $2,172M (8.6% GAAP operating margin) and adjusted operating profit of $3,292M (15.1%). Of FY25 sales, new equipment was 72% and parts & service 28%; international (incl. US export) was ~52%.

The business model has two engines. (1) Equipment — point-in-time product sales recognized at shipment, cyclical, tied to housing/construction/replacement cycles. (2) Aftermarket — the recurring, higher-margin annuity of replacement components, preventive and on-demand maintenance, digital monitoring (Abound building platform; Lynx cold-chain platform built with AWS), and modernization/upgrades over a 15-20 year installed-base life. The entire strategic thesis is to grow the aftermarket share of the mix — it is the moat and the margin.

Brands: Carrier, Viessmann, Toshiba, Automated Logic, Bryant, CIAT, Day & Night, Heil, NORESCO (HVAC); Carrier Transicold, Sensitech (transport).

Customers span residential, commercial, education, healthcare, technology, retail, hospitality, data center, and infrastructure markets, sold directly to contractors/owners and indirectly through JVs, distributors, dealers, and retail. Carrier holds interests in ~58 joint ventures (52% tied to CSA, 38% to CSAME) — the most important being the Carrier Enterprise JVs with Watsco (the dominant US HVAC distributor) and various Midea Group JVs. Equity-method JV net earnings were $229M in FY25 — a material, often-overlooked profit stream (JV investees did $18.2B of sales, $676M net income FY25 ).

Contract structure: mostly point-in-time (control transfers at shipment); installation/service contracts recognized over time (cost-to-cost). No take-or-pay; the recurring-ness is from the installed base + service contracts, not contractual minimums.

The frontier line — datacenter cooling (the reason this is in the datacenters bucket): Carrier QuantumLeap is an integrated chip-to-chiller thermal-management system combining traditional + liquid cooling with building/server management. Management targets ~$1.5B of datacenter sales in 2026, entered 2026 with ~$1B datacenter backlog, and reported datacenter orders +400% in Q4'25 and +500% in Q1'26. This is ~7% of revenue but is where the bull case lives.

Supply Chain

Carrier sits in the middle of the building-systems chain: raw-material/component inputs → Carrier manufacturing (~850 sites, ~36M sq ft ) → distribution → installed in buildings/vehicles → lifecycle aftermarket.

Upstream (inputs): steel, aluminium, copper, refrigerants, compressors, motors, electronic controls, semiconductors, and supplier-provided parts. Carrier flags raw-material and supplier dependency as a named risk — "ability of suppliers to deliver materials, parts, components and manufacturing equipment". Notably for the datacenter line, fluorosurfactants and refrigerant chemistry matter (and are the root of the AFFF legal tail — see Lens 10). Carrier does NOT manufacture fluorosurfactants — it bought them from third parties.

Midstream (Carrier itself): ~850 sites; ~34% of significant properties in CSA, 22% CSE, 26% CSAME, 10% CST; ~34% leased, ~29% in the US.

Downstream (distribution → end customer) — named stakeholders:

• Watsco, Inc. — via the Carrier Enterprise JVs, the primary US residential/light-commercial distribution channel. This is the single most important channel relationship; a Watsco destocking cycle directly hit FY25 (residential volume −9% partly on "distributor destocking").
• AHI-Carrier FZC — UAE-based JV (Middle East distribution).
• Midea Group JVs — China/Asia manufacturing + distribution.
• Independent sales reps, distributors, wholesalers, dealers, retail outlets (residential); direct to contractors/owners (commercial); transport companies + dealers (CST).
• Datacenter chain: hyperscaler/colocation operators are the end buyers; Carrier partners on liquid cooling with ZutaCore (Carrier Ventures expanded its investment April 2026 for direct-to-chip waterless cooling); QuantumLeap integrates chillers + Coolant Distribution Units (CDUs) + digital twins.

Chokepoints / single-source dependencies: (1) Watsco channel concentration in US residential — Carrier's results swing with one distributor's inventory decisions. (2) Refrigerant regulatory transitions (low-GWP mandates) force product redesign on a regulatory clock. (3) China JV exposure via Midea amid geopolitical risk. (4) Semiconductor/electronic-controls supply for digitally-enabled and datacenter products.

Competitive Advantages (moats)

Moat 1 — Installed base + aftermarket annuity (the real one). ~28% of revenue is parts & service on a multi-decade installed base. Replacement HVAC is non-discretionary (when the unit dies in July, it gets replaced), and service relationships are sticky. This is the durable, recession-resistant core. Carrier's explicit strategy is to grow this via digitally-enabled lifecycle solutions (Abound, Lynx).

Moat 2 — Brand + scale + distribution density. Carrier is the #1 HVAC brand in North America at ~17% share, ahead of Trane (10%) and Lennox (8.2%). The Carrier-name + Viessmann (premium Europe, direct-to-installer) + Toshiba (Asia) brand stack, plus the Watsco distribution density, are hard to replicate. Scale gives purchasing leverage on steel/copper/components.

Moat 3 — Switching costs in commercial/applied + controls. Building controls (Automated Logic, Abound) and large applied-HVAC installs create integration lock-in; once a chiller plant + BMS is specified, replacement and service tend to stay in-family.

Moat 4 (emerging, unproven) — Datacenter thermal integration. QuantumLeap's chip-to-chiller integration + the ZutaCore direct-to-chip stake is an attempt to build a position in AI-datacenter cooling. But this is a contested, not-yet-defensible moat — Trane (LiquidStack acquisition), Vertiv, AAON/BASX, and Asian majors are all racing here.

Bargaining power: Over suppliers — strong (scale buyer of commodity inputs). Over customers — mixed and arguably weakening in residential: the Watsco-channel dependence and distributor destocking show the channel can dictate timing; in commercial/datacenter, buyer power is higher (hyperscalers are sophisticated, multi-source). The FY25 organic decline (−1%) against a backdrop of "improved price" in commercial but volume collapse in residential/light-commercial shows pricing power is real but not enough to offset volume in a downcycle.

Segments

Carrier recast to four regional Climate Solutions segments in May 2025 (CSA, CSE, CSAME, CST); all prior periods restated. Segment profit measure is Segment Operating Profit (excludes restructuring, acquired-intangible amortization, nonoperational items).

FY2025 by segment:

Geographic mix (FY25): International (incl. US export) ~52% of sales. Q1'26 geo split: US $2,657M, Europe $1,549M, Asia-Pac $986M, Other $149M.

Segment trends and causes:

• CSA (48% of sales, the profit engine): flat-to-down — residential volume −9% (weak end-markets + Watsco destocking), light-commercial −20%, partly offset by commercial +23% (datacenter/applied demand + price). The mix shift commercial-up/residential-down is the whole story. Margin compressed 160bps to 20.5% on residential deleverage.
• CSE (23% of sales): +1% reported but organic −3%; the Viessmann (VCS) acquisition (closed Jan 2024) dominates this segment. Residential/light-commercial −5% on European weakness/regulatory uncertainty; commercial +2%. Lowest margin (8.8%) and the goodwill-risk segment (Lens 10).
• CSAME (15% of sales): −5% reported, organic −5% — China residential −12% on economic weakness; rest of region partially offsetting. Q1'26 guidance cut ~50bps on Middle East conflict + China.
• CST (13% of sales): −17% reported (CCR commercial-refrigeration divestiture) but organic +4% — containers +31% on demand + price; global truck/trailer −3%. The one segment with positive organic growth and improving margin (14.0%→15.6%).

Product-vs-service disaggregation (FY25, all ): Product $9,327M / Service $1,143M (CSA); Product $4,572M / Service $472M (CSE); Product $2,606M / Service $733M (CSAME); Product $2,668M / Service $226M (CST). Note Viessmann lifted CSE product sales sharply from FY23 ($1,722M) to FY25 ($4,572M) — the acquisition is visible here.

Phase B — Measure performance

Earnings Result (latest print — Q1 2026)

The Q1 2026 print is the single most important data point in this dossier: a sharp profit deceleration masked by an order-growth headline.

Top line: Q1'26 net sales $5,341M, +2.4% reported but −1% organic vs Q1'25 $5,218M. Beat consensus on revenue.

Profit — the problem: GAAP operating profit collapsed 59% to $259M from $629M. Net earnings to common $238M vs $412M (−42%); GAAP diluted EPS ≈ $0.28 (238/842.8 ). On a non-GAAP basis, adjusted EPS was $0.57, down 12% YoY — the gap between $0.28 GAAP and $0.57 adjusted is the $108M restructuring + $213M acquired-intangible amortization add-backs.

What drove the miss in profit:

• CSA segment operating profit −35% to $373M (margin 22.2%→14.9%) — residential volume deleverage.
• CSE −15% to $89M, hit by $86M of CSE restructuring (vs $0 prior-year quarter).
• CSAME −33% to $81M (China + Middle East).
• CST +4% to $101M — the bright spot again.
• Total segment operating profit $644M vs $893M (−28%).

Guidance & tone: Carrier reaffirmed FY26 guidance — ~$22B sales, ~$3.4B adjusted operating profit, ~$2.80 adjusted EPS, ~$2B FCF. But CSAME margin guidance cut ~50bps on Middle East conflict + China weakness. Tone: cautious on residential/China, bullish on datacenter/commercial orders.

The order-vs-earnings divergence (key tension): total orders +11%, Commercial HVAC orders +35%, datacenter orders +500% — yet current-quarter earnings fell. Orders are a forward signal (good); the P&L is a present reality (bad). The bull leans on orders; the bear leans on the P&L and cash flow.

Balance-sheet/cash flags: Q1 is seasonally cash-negative for HVAC; the 8-KForm 8-KA filing for something that happened between reports and matters enough to tell shareholders now. headline flagged cash flow falling YoY. Goodwill ticked down to $15,313M (FX). Restructuring spend ramped sharply ($108M vs $8M) — a margin headwind now, a productivity bet later.

Market reaction: Despite the order surge, the stock has gone roughly nowhere-to-down — ~$73 in late June 2026 vs a 52-week that prices in the residential drag; the order growth did NOT trigger a re-rate, telling you the market is skeptical the datacenter story offsets the core cycle.

Earnings Calls (sentiment trend)

No transcripts/ on the shelf (transcripts=0) — this lens is -grounded from the Q1'26 call and management commentary.

What management is focused on (recurring themes across recent calls):

1. Datacenter / AI cooling — the dominant new narrative. "Datacenter orders +400% (Q4'25) → +500% (Q1'26)," "$1.5B 2026 datacenter target," "$1B backlog entering 2026," QuantumLeap launches, ZutaCore. Tone here is unambiguously bullish — this is the story they want investors to buy.
2. Aftermarket / recurring revenue — persistent multi-year theme; digitally-enabled lifecycle solutions.
3. Portfolio simplification "complete" — they stopped talking about Fire & Security; the pivot phrase is now "pure-play climate and energy".
4. Carrier Energy — newer narrative on home energy management / grid flexibility.

What they're saying MORE of: "data center," "liquid cooling," "QuantumLeap," "orders" (a tell — when management pivots to orders/backlog, present earnings are usually soft). What they're saying LESS of: Viessmann synergies as a headline (now in run-rate), Fire & Security.

Sentiment shift: the tone has migrated from "transformation / deleveraging / synergy" (2024-25) to "AI datacenter growth + cautious-on-residential" (2026). The caution on China/Middle East and the reliance on the order book (vs reported profit) is the honest read: management is using the datacenter story to bridge investors over a weak residential cycle. CEO David Gitlin's open-market share purchase is a credible confidence signal.

Comps

Peer set: the global pure-play HVAC/climate names. Multiples are **forward 2026 **, dated June 2026; where I cannot source a figure I mark it n/a. Do not treat these as precise to the decimal — sources vary.

Read of the table:

• Carrier trades roughly in-line-to-slightly-cheap vs JCI/LII (~26x) but at a clear DISCOUNT to Trane (~31x). The market is paying a ~5-turn premium for Trane's cleaner growth/margin profile and its perceived datacenter leadership (LiquidStack) over Carrier's residential-drag + goodwill overhang. The entire bull case can be framed as "Carrier closes the gap to Trane."
• Carrier's EV/EBITDA (23.0) is actually the highest of the group despite the lowest P/E — because of its high net debt (~$10.4B) and goodwill-heavy balance sheet inflating EV. That is a flag: on an enterprise basis, Carrier is NOT cheap.
• Carrier's ROE (9.9%) and ROIC (7.25%) are the weakest fundamentals in the group — Trane and Lennox earn far higher returns on capital. Carrier is the lowest-quality-return name trading at a near-market multiple. That is the bear's core valuation point.
• Trailing P/E (49x ) is distorted by the FY24 discontinued-ops gain washing out of the trailing base; use forward (26x) and adjusted EPS for any real comparison.

Stock-Price Catalysts (what moves CARR >5%)

Mostly; the pattern over 2020-2026:

• 2020 spin-off from UTC (Apr 2020) — Carrier began trading independently; the post-spin re-rating (from ~$11 to multi-bagger through 2021) was the original value-creation event.
• Apr 2023 — portfolio-transformation announcement (Viessmann acquisition + Fire & Security/CCR exits). A defining strategic catalyst that reset the equity story to "pure-play HVAC."
• 2024 — divestiture completions ($10B+ gross: Access Solutions $5.0B, Industrial Fire $1.4B, CCR $679M, CRF $2.9B) and the $4.5B discontinued-ops gain that flattered FY24 net income. Proceeds funded buybacks.
• Mar 2026 — "datacenter orders +400%" (Q4'25 call) — the catalyst that introduced the AI-cooling narrative.
• Apr 30 2026 — Q1'26 print — orders +500% datacenter / +35% commercial, but GAAP OP −59%, adj EPS −12%; net reaction muted-to-negative.

What the pattern reveals: historically CARR moved on portfolio/strategic events (spin, M&A, divestitures) more than on quarterly beats. The new swing factor is datacenter order momentum — but Q1'26 showed the market now wants order growth to convert to earnings, not just bookings. The dominant macro driver is the US residential HVAC replacement cycle + rates (housing turnover, repair-vs-replace), with China/Europe macro as secondary drags. This is a rate-sensitive, housing-sensitive industrial wearing an AI-growth costume.

Phase C — Judge people & books

Management

CEO — David Gitlin (Chairman & CEO since the 2020 UTC spin; Chairman since Apr 2021; age 56). Background: President & COO of Collins Aerospace (2018-19), President of UTC Aerospace Systems (2015-18) — an aerospace operator, not a lifelong HVAC person.

1. Track record: Gitlin took Carrier from a UTC carve-out to an independent pure-play, executing a >$10B divestiture program (Fire & Security, CCR, CRF, Access) and a $14.2B Viessmann acquisition in 24 months. Operationally he delivered the spin and the transformation; the question is whether the portfolio he built earns its cost of capital.
2. Tenure & skin in the game: ~6 years as CEO; CEO made an open-market share purchase signaling confidence — a positive insider tell. (No our figures on the shelf — flagged as a gap.)
3. Capital-allocation history — the weak spot: Carrier's ROIC fell from ~18.4% (2018, under UTC) to ~7.25% (2026). The Viessmann deal added $7.6B of goodwill and a low-margin (8.8%) European segment, structurally diluting returns. Offsetting: aggressive buybacks — $12.1B authorized (raised $5B Oct 2025), 114.9M shares repurchased for $6.8B since 2021, $5.3B remaining; $2,892M repurchased in FY25 alone; dividend raised to $0.24/qtr ($0.96 annualized, 1.47% yield). Verdict: shareholder-friendly on returns of capital, but the headline M&A (Viessmann) has so far depressed ROIC — the jury is out on whether Europe + datacenter justify the price paid.
4. Red flags: (a) ROIC compression post-Viessmann. (b) Heavy reliance on non-GAAP adjustments — FY25 adjusted operating profit ($3,292M) is 52% higher than GAAP operating profit ($2,172M), driven by $856M acquired-intangible amortization + restructuring; investors must judge how "non-recurring" perennial restructuring really is. (c) New CAO (Beril Yildiz, Sept 2025) and CFO title change (Goris → EVP CFO & Strategy, Jan 2026) — leadership churn in the finance org around a complex transformation.
5. Founder vs professional manager: Professional manager / portfolio operator (aerospace pedigree). Implication: disciplined on process and capital returns, but the value creation rests on M&A integration and multiple re-rating, not founder-led product vision. For this mature-industrial stage, that's appropriate.

CFO — Patrick Goris (CFO since Nov 2020, ex-Rockwell Automation CFO; promoted to EVP CFO & Strategy Jan 2026) — a respected industrial CFO; the "& Strategy" expansion signals more M&A/portfolio work ahead.

Forensic Red Flags

Acting as a forensic analyst. Two genuine flags, several to watch.

FLAG 1 (most important) — Climate Solutions Europe goodwill, thin headroom. Total goodwill is $15,501M (Dec-31-2025) — 110% of total equity of $14,128M. Of that, CSE carries $7,808M of goodwill (the Viessmann excess). Carrier ran a quantitative impairment test on CSE (qualitative "step zero" for all others) and found fair value only ~14% above carrying value. CSE is the lowest-margin segment (8.8%) and decelerating organically (−3%), with $86M of restructuring in Q1'26 alone. Management explicitly warns: "a significant increase in the discount rate, decrease in the long-term growth rate or substantial reductions in our end markets... could have a negative impact". A European recession or a higher-for-longer rate environment could force a multi-billion-dollar CSE goodwill impairment. This is the largest single accounting risk on the balance sheet.

FLAG 2 — AFFF / PFAS litigation tail (Kidde-Fenwal). Carrier inherited Kidde-Fenwal (KFI) from the UTC separation; KFI's "National Foam" business made AFFF firefighting foam now alleged to have released PFAS. >17,000 lawsuits in a South Carolina MDL. KFI filed Chapter 11 (May 2023). Carrier entered a Settlement & Plan Support Agreement: pay $615M cash over 5 years + ~$115M net KFI asset-sale proceeds + insurance rights. It recorded a $615M liability ($565M in 2024 + $50M in 2023) and expects insurance (up to $2.4B available) to cover the cash. But the critical caveat: for remaining/non-settling claims, Carrier "is unable to assess the probability of liability or to reasonably estimate a range of possible loss... There can be no assurance that any such future exposure will not be material". As of Dec-31-2025 it has recorded zero for expected insurance recoveries. The settlement still needs Bankruptcy Court approval (disclosure statement modified twice, Oct 2025). Net: largely ring-fenced and insured, but a live, court-dependent tail with explicit "could be material" language.

WATCH items:

• Non-GAAP gap: $884M of intangible amortization + perennial restructuring inflate adjusted profit ~52% above GAAP. Quality-of-earnings caution — judge how much restructuring is truly one-time.
• Receivables/inventory vs revenue: segment assets (AR + inventory) rose to $5,035M from $4,765M while revenue FELL 3% — working capital building into a downturn (watch for channel stuffing or demand misread; consistent with the Watsco destocking narrative).
• Cash divergence: FY25 GAAP net earnings $1,587M vs OCF $2,513M — OCF exceeds earnings (healthy, driven by D&A $1,274M). No red flag here; cash conversion is solid.
• Valuation allowances rose to $1,539M — large but stable; tax-asset realizability to monitor.
• Tax rate volatility: FY25 effective tax 13.3% ($240M on $1,798M pretax) vs FY24 46.7% — driven by discrete items/divestitures. Normalize tax in any EPS model.

Regulatory findings (required sub-section):

• SEC: No Litigation Releases and no AAERs name Carrier Global (searched EDGAR EFTS, LR + AAER, 2021-06-30 to 2026-06-30). Clean SEC record.
• Item 3 Legal Proceedings (10-K): Asbestos (legacy, "not material individually or in aggregate" ) and the AFFF/PFAS MDL detailed above. AFFF is the only material litigation.
• Non-SEC enforcement (web): No material FTC/DOJ/FDA/CFPB enforcement actions or consent decrees surfaced for Carrier Global as of June 2026. The PFAS exposure is civil/product-liability (MDL), not an agency enforcement action against Carrier.
• Conclusion: No material regulatory (SEC/agency) findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 as of 2026-06-30. The one material legal contingency is the AFFF/PFAS MDL, substantially settled-and-insured but court-pending with residual tail risk.

Phase D — Project & stress-test

Forward Projection

Building bottom-up from FY25 actuals + FY26 guidance. All outputs; inputs labeled. Adjusted (non-GAAP) EPS basis to match how the company guides and the street models (GAAP will run ~$1.10-1.30 lower per share due to ~$0.95-1.05/sh of intangible amortization on ~835M shares ).

Anchors:

• FY25 adjusted operating profit $3,292M; FY25 GAAP cont-ops EPS $1.69, total $1.72.
• FY26 company guidance: ~$22B sales, $3.4B adj op profit, **$2.80 adj EPS**, ~$2B FCF; street ~$2.79-2.88.
• Buyback engine: ~$5.3B authorization remaining; FY25 reduced diluted share count from ~912M (FY24) to ~862M and Q1'26 to ~843M — ~3-4%/yr share-count reduction is a reliable EPS tailwind.

Base case (most likely):

• FY2026E adj EPS ~$2.80 — in line with guidance: flat-to-LSD organic, Riello divestiture −$0.03, restructuring savings building, ~3% buyback accretion offsetting residential drag.
• FY2027E adj EPS ~$3.15 — assumes residential cycle troughs and modestly recovers (+MSD organic), commercial/datacenter sustains double-digit, ~50bps margin recovery from restructuring run-rate, ~3% buyback.
• FY2028E adj EPS ~$3.55 — normalized HVAC recovery + datacenter scaling toward $2.5-3B, continued buyback.

Bull case: residential snaps back faster (rate cuts), datacenter hits/exceeds $1.5B→$3B trajectory at premium margin, CSE restructuring delivers, multiple re-rates toward Trane. FY26 ~$2.90, FY27 ~$3.40, FY28 ~$4.00.

Bear case: residential trough persists/deepens, China + Europe stay weak, CSE goodwill impairment (non-cash but sentiment-damaging), datacenter orders don't convert to margin (competitive pricing). FY26 ~$2.65, FY27 ~$2.75, FY28 ~$2.90 — i.e. EPS stagnates.

The valuation math: at ~$73 and ~$2.80 FY26 adj EPS = ~26x forward. For a name with ~7% ROIC and flat organic growth, 26x is a growth-priced multiple on a cyclically-stuck industrial — it only works if (a) the residential cycle inflects AND (b) datacenter re-rates the stock toward Trane's 31x. On the base case, a fair multiple of ~22-24x on FY27 ~$3.15 = ~$69-76 fair value — roughly where it trades. The stock is approximately fairly-to-fully valued; the asymmetry is not obviously favorable at this price.

Brier forecast: Skipped per --watchlist rules (only log a forecast when genuinely committed to the base case; this is a breadth-mode dossier). Candidate if promoted: "CARR FY2026 adjusted EPS >= $2.80, p≈0.55, resolves 2027-02-28."

Bull vs Bear

Bull case. Carrier is the #1 North American HVAC brand with a durable, growing aftermarket annuity (28% of revenue, recession-resistant) attached to a multi-decade installed base. It is the cheapest large HVAC name on P/E (~26x vs Trane ~31x) with a credible path to close the gap. The datacenter/AI-cooling option is real and accelerating — $1.5B 2026 target, $1B backlog, orders +500%, QuantumLeap + ZutaCore positioning Carrier in the highest-growth thermal-management niche of the decade. Secular tailwinds: electrification/heat pumps (Viessmann), decarbonization, cold-chain digitization (Lynx/AWS), and AI-datacenter buildout. Capital allocation is shareholder-friendly: $5.3B buyback remaining (~9% of cap) + growing dividend, ~3-4%/yr share shrink. When the US residential cycle inflects on rate cuts, operating leverage snaps back hard. The earnings surprise that re-rates it: residential recovery + datacenter margin conversion landing in the same year.

Bear case (2-3 permanent-impairment risks).

1. The multiple is the bet, and it's a bad bet. 26x forward (23x EV/EBITDA — the highest of the peer group) on a business earning ~7% ROIC with flat organic growth is paying a quality multiple for sub-quality returns. If datacenter doesn't re-rate it and the residential cycle stays soft, the de-rate to ~18-20x (industrial-cyclical fair value) is −25-30%.
2. CSE goodwill impairment. $7.8B of Viessmann goodwill with only ~14% headroom in a decelerating, low-margin European segment. A Euro recession or higher discount rate forces a multi-billion write-down — non-cash, but it crystallizes that the marquee acquisition destroyed value and would gut sentiment.
3. Datacenter is contested, not owned. Trane (LiquidStack), Vertiv, AAON/BASX, and Asian majors are all in liquid cooling. Carrier's $1.5B is ~7% of revenue and faces hyperscaler buyer-power + price competition. The order growth (+500%) is off a tiny base; converting bookings to durable high-margin revenue against well-capitalized rivals is unproven.

Pre-mortem (18 months out, thesis broke): It's late 2027. US residential never recovered (rates stayed high, housing turnover frozen); China and Europe stayed in recession; CSE took a ~$2B goodwill impairment; datacenter orders converted to revenue but at compressed margins because hyperscalers dual-sourced against Trane/Vertiv. Adjusted EPS stalled at ~$2.75. The 26x multiple compressed to ~19x. The stock is in the low-$50s, −30%. The buyback cushioned but couldn't offset.

Is the multiple too high? Yes, modestly — on enterprise valueEnterprise valueWhat it would cost to buy the whole company: its market value plus its debt, minus the cash you would get with it. Often very different from market cap. (23x EBITDA, highest in peer group) and on returns (7% ROIC), the equity is priced for a recovery + re-rate that is plausible but not yet in the numbers.

Contrarian view (what the market refuses to see): The market is debating "is Carrier an AI-datacenter cooling play?" The honest answer is no — it's a 93%-residential/commercial/transport HVAC cyclical with a 7% datacenter garnish, and the datacenter narrative is being used to paper over a genuinely weak core cycle. The real under-appreciated asset is the boring aftermarket annuity + the Watsco/JV equity-income stream ($229M) — but that's a "steady compounder at a fair multiple" story (own it in the high-$50s/low-$60s), not a "growth re-rate" story justifying 26x today.

Devil's Advocate (short-seller)

Dismantling the bull case as a skeptical short-seller.

• What structurally breaks the money machine? A prolonged US residential HVAC downturn. ~48% of revenue (CSA) is the profit engine; residential volume already −9% (FY25) and CSA segment OP −35% (Q1'26). HVAC replacement can be deferred in a weak economy (repair instead of replace), and new-construction is rate-hostage. This isn't a clean secular grower — it's a housing/rates cyclical.
• Where is revenue concentrated, and what shifts it? Geographically heavy on US residential via the Watsco channel — and Q1'26 proved one distributor's destocking can swing the whole segment. If Watsco's inventory discipline tightens further, Carrier's reported volumes fall regardless of end-demand.
• Why is the moat weaker than bulls think? In residential, the "moat" is brand + distribution, but Lennox, Trane, Daikin/Goodman, and the Midea-adjacent imports all compete hard; pricing power exists but didn't stop a volume collapse. The datacenter "moat" is nascent and contested — Carrier is a fast follower, not the leader (that's arguably Trane/Vertiv).
• Most dangerous underestimated competitor: Trane Technologies. Higher margins, higher ROIC, cleaner balance sheet (no $7.8B European goodwill overhang), and it bought LiquidStack outright vs Carrier's minority ZutaCore stake. Trane is winning the premium multiple and arguably the datacenter-cooling narrative. Carrier is the value-trap-risk relative to Trane's quality-compounder.
• Worst capital-allocation moves: The $14.2B Viessmann acquisition — paid up for a low-margin (8.8%) European business that immediately diluted ROIC from double digits toward 7% and loaded the balance sheet with $7.8B of goodwill carrying only 14% impairment headroom. Aggressive non-GAAP add-backs (adjusted profit 52% > GAAP) flatter the optics.
• Assumptions required for today's price: (1) US residential troughs and recovers within ~12 months; (2) datacenter scales to $2.5B+ at premium margin without margin-destroying competition; (3) no CSE goodwill impairment; (4) multiple holds at ~26x or re-rates higher. All four must roughly hold.
• If growth disappoints 20-30%: if FY27 adj EPS comes in ~$2.75 instead of ~$3.15 and the multiple compresses to ~19x, the stock is ~$52, −29%. The buyback ($5.3B) softens but doesn't save it.
• Single scenario that permanently impairs: A structural step-down in US residential HVAC demand (e.g. a multi-year housing freeze) combined with the datacenter buildout cresting earlier than expected — Carrier would be left as a low-growth, low-ROIC, goodwill-heavy industrial trading at industrial multiples (~16-18x), with the marquee European acquisition impaired. Plausibility: moderate — not a fraud or solvency risk (balance sheet is investment-grade, AFFF is insured), but a real de-rating/value-destruction path.

Management Questions (ordered by information value)

1. Datacenter is ~7% of revenue but ~100% of the growth narrative — what gross margin does the datacenter/QuantumLeap business earn today vs the corporate average, and where does it settle once hyperscalers dual-source against Trane/Vertiv?
2. CSE goodwill ($7.8B) has only ~14% fair-value headroom — at what discount rate / European growth assumption does it impair, and what's your live sensitivity?
3. Your adjusted operating profit is ~52% above GAAP, largely on restructuring + intangible amortization — how much of the perennial restructuring is genuinely non-recurring, and when does GAAP converge to adjusted?
4. ROIC fell from ~18% (2018) to ~7% — what is your explicit ROIC target and timeline, and does the Viessmann deal clear your cost of capital on your own numbers?
5. How dependent is reported US residential volume on the Watsco channel's inventory decisions, and what did Q1'26 destocking teach you about that exposure?
6. On AFFF: what is the realistic worst-case for non-settling direct claims, and what's your confidence the up-to-$2.4B insurance actually funds the $615M settlement on schedule?
7. What residential end-demand (not channel inventory) trajectory are you assuming in the ~$2.80 FY26 guide, and what breaks it?
8. Capital allocation: with ~$5.3B buyback remaining, $2B FCF, and a goodwill-heavy balance sheet — buybacks vs deleveraging vs more M&A, and what's the hurdle for the next deal?
9. Carrier Energy and Lynx/Abound are pitched as recurring-revenue engines — what is the actual recurring/ARR contribution today, and the 3-year target?
10. China residential was −12% — is that cyclical or a structural share/positioning problem given the Midea JV dynamics?
11. What is the aftermarket attach rate on the installed base, and what's the explicit plan/target to raise the 28% service mix?
12. The ZutaCore stake is a minority position vs Trane buying LiquidStack outright — why minority, and will you need to acquire liquid-cooling IP to compete?
13. With ~52% international revenue, how much of FY25's reported swing was FX vs real demand, and how are you hedged into 2026?
14. Post-Riello, is the portfolio "done," or are there further divestitures (or a CSE restructuring) on the table?
15. What single metric should investors track to know whether the datacenter thesis is converting to durable value vs just bookings — and where is it now?