
GEV · Utilities
The market has correctly identified GE Vernova as the picks-and-shovels play on the AI power demand shock, but in doing so has priced out the very margin of safety that makes owning an industrial with an unproven standalone track record rational — the insight everyone is missing is that being right about the thesis and making money are two different things when the multiple already embeds perfection.
$978.32
$500.00
The equipment-plus-services flywheel in Power is a genuinely durable moat — thirty-year service contracts with switching costs measured in safety liability, not just inconvenience — but the Wind segment remains a structural drag and ROIC, while improving sharply, is still too modest to declare this a proven compounder rather than a turnaround in progress.
The transformation from cash incinerator to free cash flow machine is real and corroborated by cash conversion quality — zero debt, a fortress balance sheet, and OCF that tracks earnings without accounting tricks signals a business that has genuinely crossed from fragile to durable.
A backlog that has grown fifty percent since the spin announcement, gas turbine orders tripling in a single quarter, and electrification compounding at twenty percent organic are not manufactured numbers — this is a business catching a multi-decade infrastructure wave at exactly the moment its operational house is in order.
At north of forty-five times EV/EBITDA for an industrial business with a five-year average ROIC that was deeply negative, the current price embeds a decade of near-flawless execution across three distinct businesses, leaving essentially no margin of safety against the multiple risks that remain genuinely open — only the most optimistic DCF scenario approaches today's price.
The risks are layered and non-trivial: the Wind segment continues burning capital while exposed to policy whiplash (the Vineyard Wind halt being the most recent wound), geopolitical entanglement in Chinese wind supply chains is unresolved, and the entire bull case rests on a gas turbine supercycle that is already widely known and therefore already priced — any deceleration in hyperscaler power demand collapses the thesis faster than the backlog implies.
GE Vernova is a high-quality business at a price that demands it become a great one. The interaction between quality and price here is the central tension: the Power segment has genuine, widening moats — thirty-year service annuities, irreplaceable combustion engineering talent, a global fleet that competes against itself for service economics — and management has demonstrated the capital discipline to triage the Wind albatross rather than defend it. That's real. But EV/EBITDA above forty for a company that spent most of the last five years destroying capital is pricing in a software-like earnings durability that heavy industrial project execution has never historically sustained. The FCF yield barely clears two percent. A genuinely exceptional industrial compounder would trade at a fraction of this multiple. The trajectory is undeniably compelling. A backlog fifty percent larger than at the spin announcement, gas turbine slot reservations with embedded incremental pricing power, and an Electrification segment that is suddenly the connective tissue for every data center megawatt in America — these are structural, not cyclical, demand drivers. Management's 2028 targets imply margin expansion that, if delivered, would mark a genuine transformation from a GE castoff into a category-defining industrial franchise. The service backlog compounding on an ever-larger installed base is the quietest but most durable part of the model. The single biggest specific risk is gas turbine order cycle normalization. The entire bull case assumes that AI infrastructure power demand forces a multi-year extension of current backlog momentum. If large language model efficiency improves faster than anticipated — or if hyperscaler capital budgets tighten — turbine orders thin, pricing power fades, and the business reverts toward its historical ROIC profile faster than a forty-five times multiple can absorb. Wind remains the open wound that can re-infect the balance sheet without warning, as the Vineyard Wind halt demonstrated in a single regulatory action.