ETN

Eaton Corporation

The highest-quality electrical infrastructure compounder in the cohort — 29.8% Electrical Americas operating margin, $19.6B record backlog, and $700K–$2M of AI-DC addressable content per MW that is 10–15× larger in dollar-volume than the power-semi BOM layer. But the AI-DC re-rate already happened. At 32× forward P/E and 27.9× EV/EBITDA, the asymmetric entry was at $250–290 in mid-2024 — that trade is gone. The thesis is real. The price is the problem.

Initial sizing: medium (2.0–3.0% NAV). The sizing is capped by the 1.1:1–1.3:1 asymmetry at current price, which does not support concentration. Medium sizing acknowledges the structural thesis is sound without pretending the current price makes the trade asymmetric. The 2:1 bar re-enters the picture below $360–380 — at that level the asymmetry clears 2.3:1.

Where to add: Below $360–380 (25–27× forward P/E). Half position on initial purchase at current; reserve other half for weakness. A 10–15% drawdown from H1 2026 earnings disappointment, Boyd noise, or AI-capex fear is the intended add point, not a reason to sell.

The most consequential structural development in the ETN thesis not in the corpus. $9.5B for $1.7B of revenue at 80%+ DC mix. Closes ETN's single largest capability gap versus Vertiv at the G2T cooling/thermal layer and reframes ETN as a grid-to-chip systems vendor rather than a switchgear company with a UPS line.

Kill 3 · Boyd integration miss. $9.5B for $1.7B of revenue is a high-stakes integration at the worst possible moment (simultaneously ramping 24 manufacturing expansion projects, suspending buybacks, navigating GOES constraints). If hyperscalers perceive Boyd's CDU capability as generic (not differentiated vs Vertiv), ETN paid $9.5B to play catch-up rather than to differentiate.

The single most important structural re-rating catalyst. Post-spin, ETN becomes a $24B Electrical + Aerospace + Thermal business at ~26% segment margins with 22–26% AI-DC revenue concentration — a fundamentally different reportable entity.

The binding constraint on Electrical Americas backlog conversion. Cleveland-Cliffs is the sole US domestic GOES producer. Global lead times are 128–144 weeks (Q2 2025). This mechanistically explains Schneider's "2028–2030 real impact" framing — the calendar mismatch is real, but it reflects the physical reality of order-to-revenue conversion gated by steel supply chains, not by demand.

flowchart LR
  classDef sev fill:#3a1a18,stroke:#b85a3c,color:#fde0d8,font-family:'Geist Mono',font-size:11px
  classDef elev fill:#3a2e1a,stroke:#c8a04a,color:#fef3d6,font-family:'Geist Mono',font-size:11px
  classDef ok fill:#1c1f28,stroke:#5a6878,color:#cdd2da,font-family:'Geist Mono',font-size:11px
  classDef anchor fill:#5a4218,stroke:#e8c46a,color:#fff5d8,font-family:'Geist Mono',font-weight:600,font-size:13px

  GOES["GOES grain-oriented
electrical steel
Cleveland-Cliffs sole US
128–144wk lead times"]:::sev
  CU["Copper
LME-tracked"]:::elev
  AL["Aluminum
global commodity"]:::ok
  GAN["GaN/SiC power semis
vendor-agnostic"]:::elev
  MEX["Mexico (Juarez · Monterrey)
switchgear · 25% S232 tail"]:::elev
  WV["Weirton expansion
+30–40% US tonnage 2H 2025"]:::elev

  ETN{{ETN}}:::anchor

  GOES --> ETN
  CU --> ETN
  AL --> ETN
  GAN --> ETN
  MEX --> ETN
  WV --> GOES

  HYP["Hyperscalers
DC orders ~3× YoY Q4 2025"]:::elev
  UTIL["Utilities
FERC Order 1920
PJM $4.8B+ pipeline"]:::ok
  AERO["Aerospace
14–15% mix"]:::ok
  IND["Industrial
20–25% mix"]:::ok

  ETN --> HYP
  ETN --> UTIL
  ETN --> AERO
  ETN --> IND
    

The calendar-mismatch trap. Order growth is real and demand is genuine, but revenue conversion is GOES-gated. The 7–9% 2026 organic growth guidance is management's acknowledgment that the physical supply chain, not demand, is the binding constraint. This is not a reason to not own ETN; it is a reason to not expect near-term earnings acceleration beyond guidance.

1. Premium-multiple compression to ABB/Siemens peers on AI-DC slowdown. Bear case is a multiple story: ETN re-rates from 27.9× EV/EBITDA to 20–24× (the ABB/Schneider/Siemens range) on any combination of AI-capex moderation + Boyd integration noise + ramp cost extension. Kill signal: two consecutive quarters of AI-DC order intake deceleration (below +30% YoY) combined with Boyd commentary flagging cost overruns or customer retention.
2. AI-capex pause. Three of Big 5 hyperscalers revising 2026 capex guidance down >15% in H1–Q3 2026 earnings calls. At 16–19% AI-DC mix (growing toward 22–26%), the equity re-rates before the revenue miss is confirmed. Utility T&D and industrial provide a genuine floor; the multiple compresses fast.
3. Boyd integration miss. $9.5B for $1.7B of revenue at the worst possible moment (24 manufacturing expansion projects, suspended buyback, GOES constraints). Kill signal: Boyd-attributable revenue running 25%+ below management's $1.7B FY26 target by Q3 2026, or two consecutive quarters of "integration challenges" or "customer retention" language in cooling segment.

Mexico tariff note. A 25% Section 232 tariff on Mexico-origin electrical equipment (15–20% probability over 24 months) is a kill criterion for full-size position but not for probe sizing. If triggered, reduce to one-third position and reassess based on ETN's pass-through execution over the following two quarters.