Initiated 2026-10-02 · Price $250.60 (as of 2026-10-02 intraday, stockanalysis.com) · Mkt cap $96.5bn basic / $98.4bn on 392.7mm diluted · Industrials / electrical and thermal equipment (data-centre infrastructure) · Model: verified

Rating: Sell — Conviction: High

Vertiv is executing well. FY2026 guidance is $14.0bn of sales, +37%, the year-end backlog was $15.0bn and investment-grade ratings arrived in February. The price already discounts more than that. At $250.60 the stock trades at ~46x FY2025 EBITDA and ~29x FY2026E on the model's own basis, which is roughly where the bull case lands. The base case of a backlog-fuelled 2027 followed by normal capital-goods fade is worth about half the price.

Conviction tests (3a-v-c): T1 pass (Base −52.0% at +1pp, −45.7% at −1pp; Sell at both; Bull −11.0% / +3.0%) · T2 pass (Gordon −67%, exit-multiple −31%; Sell on either) · T3 pass · T4 pass

The rule's High is about how robust the call is, not about timing. AI-capex momentum can keep the multiple high for longer than the fundamentals justify.

Model value range vs price
Bear $38.43Base $127.63Bull $238.95Price $250.60

Business overview

Vertiv sells the power and cooling infrastructure that sits between the grid and the server: UPS, switchgear, busway and power distribution; air and liquid thermal management (CDUs, chillers, rear-door heat exchangers); integrated modules; and a services business that was ~18% of FY2025 sales. Customers are hyperscalers, colocation operators, enterprises and telecom. It reports by region: Americas $6.39bn (+41.9% in FY2025, 26.8% adjusted operating margin), APAC $2.02bn and EMEA $1.82bn. EMEA was −2.1% organic with margins down 3.8pp. Three variables drive earnings: AI data-centre capex (order intake and backlog conversion), price and mix versus copper, steel and labour costs, and fixed-cost leverage on new capacity. Customer deposits are a fourth, cash-only driver: deferred revenue rose from $1.06bn to $3.63bn in eighteen months.

Competition

In the thermal-management market, Vertiv holds ~20% share. Johnson Controls has 11%, Stulz 7%, and Trane, Schneider and Modine 6% each (Dell'Oro/Omdia data as cited by introl.com; roughly 2024 vintage). In data-centre electrical equipment, Schneider leads at ~21%, ahead of Vertiv (~17%), Eaton (~16%) and ABB (~9%). Pressure would show up first in liquid cooling. Its share is unsettled and its entrants are well capitalised (Schneider via Motivair, Modine, nVent, Trane), and price there falls fastest as capacity catches up with demand. Eaton is the closest listed comparable: 28.7x trailing EV/EBITDA and 29.0x forward P/E, against Vertiv's 35.5x and 31.8x on the same aggregator definitions.

Bull case

  1. Backlog becomes a multi-year build. The $15.0bn backlog (+109% y/y, Q4 2025 book-to-bill ~2.9x) carries FY2027 and AI campus programmes keep re-ordering, taking revenue to ~$29bn by FY2030. Plays out if hyperscaler and neocloud capex keeps growing through 2028. Model: rev_growth (bull)
  2. Content and mix lift margins. Liquid cooling, 800V DC power trains and services mix raise EBITDA margin from 20.9% (FY2025) to ~28.5%. Q3 is already guided at 24–25% adjusted operating margin. Plays out if pricing holds as capacity comes online. Model: ebitda_margin (bull)
  3. The market keeps paying a premium. Investment-grade, net-cash, ~20% EBITDA growth: the stock holds today's peer-level 26x exit at a peer-beta discount rate. Plays out if the AI build is viewed as a decade-long cycle. Model: exit_ev_ebitda, wacc (bull)

Bear case

  1. Pull-forward, then digestion. Q4 2025's 2.9x book-to-bill pulled orders ahead. Since then the company has stopped disclosing quarterly orders and backlog. FY2026 is locked, but FY2027 grows only 8%, then revenue falls 10% and 6% as hyperscalers digest capacity, as in earlier telecom and data-centre build cycles. Plays out if AI capex guidance flattens in 2027. Model: rev_growth (bear)
  2. Price competition on new capacity. Schneider, Eaton, ABB and the liquid-cooling entrants all added capacity for the boom. In a slowdown, price and fixed-cost deleverage take EBITDA margin back to ~18%. Plays out if lead times shorten and order growth stalls. Model: ebitda_margin (bear)
  3. Multiple normalisation. A 46x trailing multiple gives way to cyclical capital-goods pricing (12x exit) and a beta-2.07 discount rate. Plays out in any growth scare. Model: exit_ev_ebitda, wacc (bear)
  4. The working-capital tailwind reverses. Operating NWC went from 15.0% of revenue in FY2022 to about −5% at June 2026 on customer advances. As the backlog converts, or as orders gap, that cash flows back out. Model: nwc_pct_rev (bear)

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
EV/EBITDA, model basis (mkt cap − cash + debt + contingent consideration, ex-leases) 46.1x FY2025; ~28.9x FY2026E n/a (unverified) — Model inputs; q425rel, q2rel, 2026-06-30
EV/EBITDA, trailing (aggregator, may include leases) 35.5x n/a (unverified) ETN 28.7x · NVT 26.9x · MOD 23.6x stockanalysis.com, 2026-10-02
Forward P/E (aggregator consensus) 31.8x n/a (unverified) ETN 29.0x · NVT 28.9x · MOD 21.6x stockanalysis.com, 2026-10-02
EBITDA margin (GAAP op. profit + D&A) 22.0% H1 2026 9.2% / 16.7% / 20.5% / 20.9% FY2022–25 n/a q423rel, q425rel, q2rel

Model-implied value range (from model-summary.json; generic module, Gordon-growth DCF and exit-EV/EBITDA DCF, midpoints): Bear $38.43 · Base $127.63 · Bull $238.95 per share, i.e. implied returns of −84.7% / −49.1% / −4.7% vs $250.60. These ranges show how the bull and bear drivers translate into value; they are not price targets. The price sits slightly above the Bull midpoint, so the market is pricing roughly ~$29bn of FY2030 revenue at a 28.5% margin, or a premium multiple that never fades. The Base 18x exit is a ~61% de-rating from today's same-basis 46.1x FY2025 multiple (~38% from ~28.9x FY2026E). The two methods differ widely (Base Gordon $82.18, exit $173.07). The exit method embeds the higher terminal multiple, and even it is 31% below the price. No tail sensitivity was run: no customer concentration is disclosed (n/a), and the bear is already a break, not a plateau.

Balance sheet: not meaningful (net cash). $3,110.6mm of cash and short-term investments against $2,967.2mm of notes at 2026-06-30; net leverage −0.07x FY2025 EBITDA, gross 1.4x; EBITDA/interest ~14x. Nearest material maturity is the $850mm 4.125% secured notes in 2028. The March 2026 refinancing moved the rest to 2036–2066 unsecured notes (4.85%–5.95%), with a new $2.5bn revolver. Ratings: Moody's Baa3, S&P BBB-, Fitch BBB- (upgraded February 2026).

Model note: Verified: LibreOffice recalculation matched all 3,207 formula cells. No unverified inputs and no assumptions without basis. No scenario consistency CHECKs. The coupon on the $17.2mm of "other notes" is an estimate (~3%) and moves only EPS. Base Y1 EPS $5.77 is 1.7% below the GAAP guidance midpoint of $5.87. The difference is net interest and tax rate; adjusted EPS guidance ($6.65–6.75) excludes intangible amortization. The WACC of 13.6% equals the CAPM arithmetic in its basis. Beta is 1.73, a blend of the stock's own 2.07 and the peer mean of 1.39, which is a deliberate haircut to the stock's own beta. The FY2026 starting NWC of 0% of revenue is above June 2026's −5%. It assumes some of the deposits unwind in H2.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − Beta ~2.1 AI-infrastructure name; de-risking compresses the multiple, backlog protects revenue Med −$20.21
S2 Slow bear / recession − Colocation/enterprise projects deferred, capex slows, fixed-cost deleverage. Transitory, so milder than the permanent Base − Bear gap ($89.20) High −$55.04
S3 Rapid rate shock − Net cash and fixed-rate notes, so the hit is duration: discount rate and growth multiple; developer financing costs rise Med −$20.28
S4 Slow rate grind − Same duration channel, grinding Low −$8.71
S5 Soft-landing cuts + Lower discount rate, budgets intact, cheaper project finance Med +$14.79
S6 Recession-driven cuts − Capex cuts and cyclical de-rating outweigh the lower rate High −$32.79
S7a Credit liquidity shock − No funding need (IG, net cash); forced selling of crowded AI names Med −$12.13
S7b Slow default cycle − Debt-funded neoclouds and colocation developers lose financing Med −$21.97
S8 Stagflation − Copper, steel and labour inflation on fixed-price backlog; higher real rates Med −$21.26
S9a Dollar spike − ~38% of sales in APAC/EMEA: translation and competitiveness drag Low −$1.98
S9b Dollar slide + Mirror of S9a Low +$1.98
S10 Melt-up + High-beta AI leader; momentum extends the multiple Med +$20.21
S11 Energy supply shock − Freight and materials cost, risk-off multiple; on-site power demand a small offset Low −$5.56
S12 Mega-cap/AI derating − The direct channel: multiple unwinds and hyperscaler capex trims hit orders once the backlog runs off High −$56.22

Currently active/on watch per the playbook: S3 partially active (state.md; the 2026-10-01 macro log has both pace legs met at 10Y 5.24%); S8, S10, S11 on watch.

Model value change vs Base, by scenario
S12 Mega-cap/AI derating−$56.22S2 Slow bear / recession−$55.04S6 Recession-driven cuts−$32.79S7b Slow default cycle−$21.97S8 Stagflation−$21.26S3 Rapid rate shock−$20.28S1 Fast equity crash−$20.21S7a Credit liquidity shock−$12.13S4 Slow rate grind−$8.71S11 Energy supply shock−$5.56S9a Dollar spike−$1.98S9b Dollar slide+$1.98S5 Soft-landing cuts+$14.79S10 Melt-up+$20.21

What would change the call

Upgrades if: the price falls toward the Base range without a change in fundamentals; or FY2027 guidance (likely February 2027) implies growth of 30% or more with EBITDA margin above 27%, along with renewed order/backlog disclosure showing book-to-bill still above 1.2x. Downgrades if: n/a (already Sell); the call strengthens if 2027 hyperscaler capex guides flat or deferred revenue falls while sales grow.

Watch items

Sources