
Eaton Corporation
Diversified industrial OEM. Designs and manufactures electrical power-distribution, grid and data-center electrification hardware (Electrical Americas + Electrical Global, ~70% of sales), aerospace systems, and vehicle/eMobility powertrain; sold via distributors and direct to utilities, hyperscalers, OEMs and contractors. Revenue mixes large-project orders, catalog product and aftermarket, and is increasingly backlog-driven (~$19.6B total backlog, Q4 FY25).
The thesis on this name
State of AI Compute
Best risk-adjusted gear-toll exposure — 37.6% group gross, diversified FCF, the lowest multiple of the US electrical trio, Mobility spin as a margin catalyst; the verify's inverted ranking makes ETN (and Schneider) the PRIMARY Phase-C grid entries (real margin-of-safety, base-rate support) vs the chase-priced VRT/GEV.
State of AI Compute
Long Eaton (ETN) over a 12–24mo horizon as the best risk-adjusted "gear-toll" on AI/data-center electrification — owning the lowest-multiple, most-diversified-FCF leg of the US electrical trio (vs chase-priced VRT/GEV) into a Mobility-spinoff re-rate, where the edge is structural (irreplaceable grid/switchgear positio…
State of AI Compute
Lowest-multiple, most-diversified electrical/grid toll on AI-datacenter electrification; the one name clipped as a real core. Enter 2/3 now ($165k), final 1/3 only on a Q2 orders beat; hard 8% cap on the ETN+VRT+GEV ele…
State of Data-Center Power
The lowest-multiple, most-diversified leg of the US electrical trio and the cleanest own-the-moat toll on datacenter electrification. Q1 2026 (fact) was a record: Electrical Americas 12-mo rolling orders +42%, datacenter orders ~+240% YoY, total backlog $14.5B (+48% Electrical). The grid/switchgear position is irreplaceable (4-5yr transformer lead times protect pricing) and FCF is real and diversified across aerospace/vehicle, so it is far less of a pure AI-capex bet than VRT/GEV. Margin slipped 120bp to 22.7% on the cost of scaling capacity — a quality problem, not a demand problem. Own on weakness as the core grid-toll; the Mobility spin is a margin re-rate optionality on top.
State of Data-Center Power
The electrical-equipment toll on every datacenter watt — transformers/switchgear are the slowest physical bottleneck (2-5yr lead times), and Eaton owns the NA installed base.
State of Data-Center Power
NA installed-base leader in transformers/switchgear, the physically slowest bottleneck (2-5yr lead); diversified base cushions the theme.
Earnings, margins, COGS & capex
Eaton has re-rated into a secular-growth compounder: FY25 revenue $27.4B (+10%), record FY25 FCF $3.6B, and a record Q1 FY26 ($7.5B sales, +17%; adj EPS $2.81). Growth is led by Electrical Americas, where data-center demand is driving backlog to records ($13.2B, +31% YoY) and data-center orders up ~200% in Q4 FY25. The near-term watch item is margin: Q1 FY26 segment margin slipped 120bps to 22.7% on input-cost inflation and ramp costs to serve 30%+ Electrical Americas volume growth, even as management targets a 30% Electrical Americas margin medium-term.
Income statement — where each revenue dollar goes
% of revenueOf every $1 of revenue, ~67¢ is cost of goods and ~10¢ operating expense, leaving ~23¢ of operating profit (~15¢ net).
Revenue trend
Margins
down (-120bps YoY on input cost + ramp)
up vs FY24, record full-year
down YoY; mgmt targets 30% medium-term
up
up (FY25 FCF $3.6B record)
COGS structure
COGS is dominated by raw materials and components - copper, aluminum, steel, electronics/semiconductors and resins/insulation - plus direct manufacturing labor across ~global plants and inbound logistics. Margin is gated by metal-price inflation, tariffs on imported inputs, and the cost to ramp new US capacity fast enough to serve data-center demand; aftermarket and software-attached revenue carry richer margins than base hardware.
Capex
Capex ~3-4% of sales (~$0.9B FY25, rising). It funds US manufacturing capacity expansion (switchgear, transformers, busway, breakers) to convert a record backlog, automation, and footprint additions tied to data-center and grid demand - a deliberate growth capex cycle, not maintenance.
Latest earnings
Clean double beat and a raise. Record sales $8,531M (+21%, 14% organic) beat consensus of ~$8.16B; adjusted EPS $3.15 (a Q2 record) beat the ~$3.07-3.08 consensus by ~2.3%; segment margins of 23.1% came in 10bps above the guidance high end. The stock rose 7.3% to $415.20 on the day. The blemish is GAAP: diluted EPS FELL to $2.11 from $2.51, on $255M of intangible amortization and $201M of net interest tied to the acquisition programme.
Raised for the second consecutive quarter. FY26 organic growth 11-13% (up 200bps at the midpoint from 9-11%, which had itself been raised from 8%); segment margins 24.1-24.5%; adjusted EPS $13.40-$13.60 (+12% at the midpoint over FY25's $12.07), up from $13.05-$13.50; GAAP EPS $10.36-$10.56, essentially FLAT (+0.1% at the midpoint) versus FY25's $10.45. Q3 FY26: organic growth 13.5-15.5%, segment margins 24.6-25.0%, adjusted EPS $3.46-$3.56, GAAP EPS $2.77-$2.87. The Mobility RMT with Dana is expected to close in Q1 2027.
- Q2 net sales
- $8,531M record (+21%; 14% organic + 7% acquisitions)
- Total backlog
- ~$24.1B at 30 Jun 2026 (71% deliverable within 12 months), up from $19.6B at Q4 FY25
- Electrical Americas
- Sales $3,951M (+18% organic); backlog $15,175M (+33%); orders +41% organic; book-to-bill 1.3
- Electrical Global
- Sales $2,517M (+44%, of which 25pts Boyd Thermal); backlog $3,602M (+103%, +54% organic); orders +33% organic
- Aerospace
- Sales $1,222M (+13%); backlog $5,164M (+28%); orders +17% organic; margin 22.8% (+60bps)
- Electrical Americas margin
- 27.5%, +190bps sequentially from 25.6% but -197bps YoY from 29.5%; 30% target unchanged
- Segment margin
- 23.1% (above guidance high end, -80bps YoY); FY26 guide 24.1-24.5% vs 22.9% in 1H26
- Gross margin
- 33.5%, down ~350bps YoY (filed, not estimated)
- Adjusted vs GAAP EPS
- $3.15 record vs $2.11 GAAP (down 16% YoY); $1.04 of add-backs vs $0.44 a year ago
- Free cash flow
- $874M (+22%), 10.2% margin
- Net debt
- ~$19.9B, roughly double the prior estimate, after $11.1B of 1H26 acquisitions
- Mobility RMT
- Dana deal signed 10 Jun 2026; Mobility valued ~$5.1B; ~$1.1B cash to Eaton; close expected Q1 2027
Growth drivers
- AI data-center electrification — power distribution, switchgear, busway and 'gray-space' content per MW; data-center orders +~200% (Q4 FY25), DC revenue +~40% YoY
- US grid modernization, utility capex, electrification and reshoring/megaprojects (IRA-era industrial buildout)
- Record backlog conversion — $19.6B total, Electrical Americas $13.2B (+31% YoY) gives multi-year revenue visibility
- Aerospace recovery - commercial OE/aftermarket up, backlog +16% YoY, defense
- Bolt-on M&A in electrical/data-center (e.g. Fibrebond, Resilient Power) adding ~4pts to growth
- Mobility spin-off (Vehicle + eMobility, ~Q1 2027) re-rating the RemainCo to a higher-margin, faster-growing Electrical + Aerospace pure-play
Reported financials — SEC EDGAR
Audited GAAP figures pulled from SEC filings · latest filing 2026-02-26. The audited primary-source spine — not financial advice.
Revenue — annual (GAAP)
Margins & balance sheet — FY’25
Bull & bear
The quarter answered the two open questions in the thesis at once: orders ACCELERATED (Electrical Americas +41% organic twelve-month rolling, backlog $24.1B) and the margin dip proved to be timing, with Electrical Americas up 190bps sequentially and segment margins above the guidance high end. Add a signed $5.1B Dana RMT with $1.1B of cash back, and the two things the bear case leaned on both broke the bear's way.
- The margin debate resolved in the bull's favour: segment margins of 23.1% beat the guidance high end and Electrical Americas inflected +190bps sequentially to 27.5% - exactly the 'the dip is timing, the leverage is structural' claim, now with a Q3 guide of 24.6-25.0% behind it.
- Demand accelerated rather than lapping out: organic growth of 14% beat the guide and accelerated from 10% in Q1, and the FY26 organic guide was raised to 11-13% - the second raise this year. Q3 is guided to 13.5-15.5% organic.
- Orders and backlog got better, not worse, against hard comps: twelve-month rolling organic orders +41% Electrical Americas, +33% Electrical Global, +17% Aerospace, with book-to-bill of 1.3 / 1.1 / 1.2 and total backlog ~$24.1B (71% deliverable inside twelve months). This is the single strongest refutation of the 'second derivative' bear point.
- The Mobility exit is now contracted, not aspirational: a definitive June 10 RMT with Dana values Mobility at ~$5.1B, hands Eaton ~$1.1B of cash (explicitly available for debt repayment), leaves Eaton holders with at least 50.1% of a >$10B combined company, and is stated to be immediately accretive to organic growth and margins on closing.
- Boyd Thermal is already earning its place: $524M of sales and $121M of segment operating profit from mid-March to June-end is a 23.1% margin, at the corporate average from day one, and it took Electrical Global backlog up 103% - it buys direct exposure to data-center liquid cooling, the fastest-growing slice of the gray-space build.
- Aerospace is quietly compounding underneath the data-center story: +13% sales, margin +60bps to 22.8%, backlog +28% and book-to-bill 1.2 - the diversification that distinguishes Eaton from the pure-play AI-power names is intact and improving.
Eaton stopped being a modestly-levered diversified compounder this half. It borrowed roughly $11B to buy Boyd Thermal and Ultra PCS, taking net debt to ~$19.9B, tripling interest expense, and pushing goodwill plus intangibles above total equity - all to concentrate further into the AI data-center cycle. GAAP EPS is now FALLING and is guided flat for the year, gross margin lost 350bps, and the full-year margin guide needs an aggressive second-half ramp.
- The balance sheet changed character: net debt ~$19.9B versus the ~$8-9B previously carried, long-term debt from $8,758M to $18,509M, and net interest expense from $71M to $201M in a single year. The 'investment-grade, sub-1.5x, continuous bolt-on M&A' framing no longer describes this company - $9.55B for Boyd Thermal is not a bolt-on.
- GAAP earnings are going backwards while the adjusted number sets records: diluted EPS fell 16% to $2.11, FY26 GAAP EPS is guided essentially flat (+0.1%) against +12% adjusted, and the add-back bridge widened from $0.44 to $1.04 a share. Goodwill $20.2B plus intangibles $12.6B now exceed total equity of $20.3B - a real impairment exposure if data-center demand normalises.
- Gross margin fell ~350bps YoY to 33.5% (filed, not estimated) and 1H26 segment margin of 22.9% sits far below the 24.1-24.5% full-year guide, which implies roughly 25.2-25.9% in 2H - a demanding ramp with no margin for a cost or tariff surprise.
- Electrical Americas margin, the single most-watched number, is still DOWN ~197bps YoY at 27.5% and remains 250bps short of the 30% target despite the sequential improvement - the sequential inflection is real but the year-over-year direction has not turned yet.
- Concentration into the AI data-center cycle deepened rather than diversifying: Boyd Thermal puts more of the company into liquid cooling just as the entire industry (Schneider, ABB, Siemens, Vertiv, GEV) expands capacity, and the RMT removes Mobility, a genuinely counter-cyclical cash stream, from the portfolio.
- The RMT still has to close: a Q1 2027 completion depends on the exchange offer, regulatory clearance and Dana shareholder mechanics, and Eaton is carrying peak leverage into that window. At ~31x FY26 adjusted EPS and ~40x GAAP, the stock is not priced for the deal slipping or the 2H margin ramp missing.
What it is worth
Forward-PE comps vs the electrical trio + reverse-DCF sanity check. ETN ~30x FY26 adj EPS (guide $13.05-$13.50) sits at the low end of the AI-power complex (GEV ~71x fwd; VRT richly priced), so the bull case is a re-rate toward peers on the Mobility-spin + margin inflection; the bear case is a de-rate toward the industrial median (~21x) on an order/margin disappointment.
~$300-340 (de-rate to ~22-24x on order normalization / margin miss)
~$450-460 (hold ~30-32x x FY26-27 EPS; consensus 12-mo target ~$452)
~$520-590 (re-rate to mid-30s x on RemainCo + raised EPS; sell-side high targets ~$590)
At ~$420 / ~30x FY26 EPS the market already prices durable double-digit growth + the 30% Electrical Americas margin; the re-rate optionality is the spin, the risk is the rich base.
SWOT
Strengths
- Diversified, backlog-rich franchise — ~70% Electrical + Aerospace gives the most balanced, lowest-cyclicality FCF of the US 'electrical trio' (vs pure-plays VRT/GEV)
- Record visibility — $19.6B total backlog, Electrical Americas $13.2B (+31% YoY), multi-year revenue coverage from data-center + grid demand
- Best-in-class cash generation — FY25 FCF $3.6B record (~13% margin), funding capex, M&A and a 60+ year dividend without leverage strain
- Scale + installed base in switchgear/distribution that hyperscalers and utilities standardize on; aftermarket pull-through
- Investment-grade balance sheet (net-debt/EBITDA <1.5x) enabling continuous bolt-on M&A (Fibrebond, Resilient Power, etc.)
Weaknesses
- Margin pressure now — Q1 FY26 segment margin -120bps to 22.7% on input-cost inflation and ramp costs; Electrical Americas at 25.6% vs 30% target
- Cyclical/lower-growth Vehicle + eMobility drag (Vehicle sales down ~8% in periods of FY25) - the reason for the spin
- Exposure to copper/aluminum/steel and tariff inflation that compresses near-term margin even as volume grows
- Execution risk converting a record backlog fast enough - capacity ramp is itself a cost headwind
- Premium valuation (~30x fwd PE) leaves little room for an order-growth deceleration
Opportunities
- AI/data-center electrification multi-year capex supercycle — rising electrical content per MW, 'gray-space' and grid interconnect
- Mobility spin-off (~Q1 2027) as a re-rating catalyst — a cleaner, higher-margin Electrical+Aerospace RemainCo could command a higher multiple
- US grid modernization, reshoring megaprojects and utility capex independent of the data-center cycle
- Software/services + electrification attach (microgrids, energy management) raising margin mix
- Continued accretive bolt-on M&A in data-center and grid niches
Threats
- Data-center order normalization or a hyperscaler capex air-pocket (the orders are +200% off a hot base; comps get hard)
- Competitive capacity additions (Schneider, ABB, Siemens, GEV, VRT) eroding pricing as the whole industry expands
- Sustained metal/tariff inflation outrunning price/cost recovery
- Spin-off execution/dis-synergies or a value-destructive separation if poorly timed
- Macro/rate-driven slowdown in non-DC commercial construction and a beta de-rating of the whole AI-power complex
Moats, dependencies & bottlenecks
Moats
Installed base + specification lock-in (switchgear/distribution standardized by utilities, hyperscalers, contractors) Long design-in cycles, code/safety certification and aftermarket pull-through create high switching costs across a multi-decade installed base.
Global plant network and procurement scale let Eaton convert a record backlog and absorb input inflation better than sub-scale rivals; capacity is itself a moat when lead times are long.
Most balanced FCF of the trio; smooths single-end-market shocks - the explicit reason it's the lower-beta way to own AI-power.
EPC contractors) Entrenched distributor relationships and project specs, but contestable by Schneider/ABB/Siemens with comparable channel reach.
In data-center and aerospace, proven reliability and certification are buying criteria; reputation compounds but is shared with a few credible peers.
Dependencies
Amazon, Google, Meta, Oracle, colos) The marginal growth driver; a capex digestion or AI-spend pause hits the fastest-growing, highest-multiple part of the story.
copper, aluminum, steel, electronics/semiconductors, resins Direct COGS driver; metal/tariff inflation is the current margin headwind (Q1 FY26 -120bps).
Large, durable but macro/rate-sensitive demand pool beyond data centers.
Tariffs raise input cost; codes/certification are a barrier that cuts both ways (moat + compliance burden).
Aerospace segment recovery depends on commercial OE/aftermarket and defense demand.
Advantages
- Lowest-multiple, most-diversified FCF leg of the US electrical trio - lower-beta AI-power exposure than VRT/GEV
- Record backlog ($19.6B) and order book gives multi-year visibility competitors with thinner books lack
- Record FCF ($3.6B FY25) self-funds growth capex, bolt-on M&A and a long-tenured dividend
- Investment-grade balance sheet enabling continuous accretive M&A
- Mobility spin-off optionality - a structural re-rate path the pure-plays don't have
- Installed base + specification lock-in across mission-critical power and aerospace
Weaknesses
- Near-term margin compression (segment -120bps to 22.7% in Q1 FY26) on input cost + ramp
- Electrical Americas at 25.6% vs a 30% target - gap must close to justify the multiple
- Vehicle/eMobility is the low-growth, lower-margin drag being spun out (sales declines in FY25)
- Heavy exposure to metal/tariff inflation in COGS
- Premium ~30x forward valuation with little margin for an order-growth miss
- Growth capex spent into a cycle that could cool, pressuring returns if demand normalizes
Bottlenecks
- Manufacturing capacity / lead times — converting the $19.6B backlog requires fast US capacity ramp; the ramp is itself a margin cost
- Skilled labor and supply-chain availability for switchgear, transformers and electronics
- Copper/aluminum/steel and semiconductor input availability and price
- Engineering/certification throughput for new data-center and grid product specs
- Price/cost recovery timing - passing tariff/metal inflation through faster than it lands
Top signals & trends
Top signals
Backlog $13.2B (+31% YoY) and DC orders +~200% (Q4 FY25) are the core thesis; watch for any sequential deceleration as comps harden.
25.6% in Q1 FY26 and falling YoY on ramp/input cost; the re-rate needs this to inflect back up - the single most-watched number.
Targeted ~Q1 2027, tax-free; a clean separation and attractive RemainCo margin profile is the re-rating catalyst.
Direct margin driver; relief would let price/cost turn positive, sustained inflation extends the margin headwind.
Eaton's data-center demand is a derivative of hyperscaler AI capex; any guide-down there is the leading bearish tell.
Already raised to ~10% from 8% at Q1; further raises confirm acceleration, a cut would break the compounder narrative.
Trends
Structural multi-year demand for switchgear, busway, distribution and grid interconnect; the dominant growth driver.
Durable non-DC demand pool (utility capex, IRA-era industrial buildout) diversifying the growth.
Compresses near-term margin; the current debate on the print.
Meets demand but risks pricing erosion mid-cycle as Schneider/ABB/Siemens/GEV/VRT all add capacity.
Sharpens RemainCo to higher-growth Electrical + Aerospace; potential multiple re-rate but adds 2027 execution risk.
Secular but currently a drag and being spun out; matters less to RemainCo post-2027.
Ecosystem & competitor graph
Suppliers feed the company; customers pull from it. Line thickness shows the strength of each tie (supply-chain dependency, customer earnings contribution). Hover to isolate a tie.
Freeport-McMoRan, Southern Copper, Nucor, Alcoa) Base-metal inputs - the dominant raw-material cost in switchgear, busway and transformers.
Semiconductor / power-electronics suppliers (e.g. ON Semiconductor, Texas Instruments, Infineon) Power semis and controls for drives, breakers and energy-management electronics.
Analog/embedded components used across Eaton's electrical and controls products.
Steel for enclosures, structural and switchgear fabrication.
Amphenol, TE Connectivity) Connectors and interconnect for power-distribution and aerospace systems.
Hyperscaler data-center buildout - power distribution and electrical infrastructure demand.
Hyperscaler data-center electrification customer driving Electrical Americas growth.
Hyperscaler AI/data-center capex - electrical gear demand.
AI data-center buildout customer; gray-space electrical content.
Aerospace segment OEM customer (fuel, hydraulics, electrical systems).
NextEra, Quanta-built projects, Digital Realty, Equinix) Grid-modernization and colocation data-center demand for distribution + interconnect.
Paris/EPA-listed (SU.PA, ADR SBGSY). The closest scaled global peer in power distribution + data-center electrical; deep channel and grey/white-space content.
Global electrification + switchgear/breakers competitor with strong data-center and grid exposure; direct rival in Electrical Global.
Data-center power + thermal pure-play; the 'chase-priced' high-beta way to own the same AI-power demand - more concentrated than ETN.
Grid/electrification + power pure-play trading ~71x fwd; richly priced peer leg of the trio, overlaps in grid/transmission gear.
US electrical + utility solutions benefiting from the same data-center/grid demand; overlaps in distribution and utility gear.
Frankfurt-listed (ENR.DE / SIE.DE; ADRs SMEGF/SIEGY). Grid, transformer and switchgear competitor with scale in transmission and data-center power.