S&P Put Meta on Negative Outlook: a 1.0x Leverage Trigger Now Governs the $225 Billion AI Capex Plan

S&P Global Ratings revised the outlook on Meta's AA− issuer credit rating to negative from stable on Oct. 30, 2026, and the trigger is not a missed payment. The agency expects free operating cash flow deficits of $22 billion in 2026, $71 billion in 2027 and about $39 billion in 2028, and says adjusted leverage could exceed 1.0x — the level that puts the AA− rating itself under review — as early as 2027, against about 0.1x in 2025. The warning is not that the company cannot pay; it is that the obligations backing a $135–145 billion annual build-out land before the cash does. Turns the AI capex story into a credit story with a deadline.

Almost none of the leverage being marked sits in reported debt. The third-quarter filing disclosed $355.73 billion of lease obligations that have not yet commenced — starting between late 2026 and 2036, with terms as long as 30 years — plus $454.52 billion of non-cancellable commitments, mostly third-party cloud capacity, servers and networking gear, data-center energy service agreements and hardware, of which $45.85 billion falls due in 2026 and $121.41 billion in 2027. Total future lease and contract obligations come to about $825 billion. Non-cancellable commitments jumped roughly $105 billion in a single quarter, from $349.31 billion at June 30 to $454.52 billion at Sept. 30, and S&P said it updated its forecast for the increased future lease obligations and now expects credit metrics to weaken beyond 2026. That single sentence is the mechanism that converts lease commitments into adjusted debt inside the rating model.

The plan is real money and it is not the money that repairs the ratio. Meta priced $35 billion on Oct. 27 — $25.0 billion of Class A stock at $625.00 a share sold forward to Morgan Stanley and Goldman Sachs, and $12.5 billion of depositary shares representing 6.50% mandatory convertible preferred — its first equity offering since 2013, and plans a $60 billion at-the-market program tied to a new Class C share class expected to trade in the first quarter of 2027. None of it changes what S&P is testing: the raise buys time and adds shareholders, not headroom. The raise buys time and adds shareholders, not headroom.

Nvidia put $3 billion into Meta's equity on Oct. 27 and the two companies deepened their AI partnership, including work on Meta's Muse agent; Broadcom's custom-chip deal with Meta runs through 2029; SK hynix is shipping HBM4 into the same accelerators. Meanwhile the asset owners and arrangers — BlackRock-managed funds, Blue Owl Capital, Vistra's nuclear plants — signed long-dated contracts that get paid whether Meta spends $145 billion or $225 billion. A negative outlook does not slow a single order. It changes who finances the customers, and it makes each supplier's order book depend on a customer's credit as well as a customer's budget. Makes every supplier's order book depend on its customer's credit.

Meta is not the first company to run this, and the outcome is already on the tape. S&P assigned Oracle a negative outlook in July 2025 and downgraded it to BBB− on July 9, 2026, one notch above junk; the shares hit a one-year low on the news. Oracle's own trailing numbers explain the action: $75.7 billion of capital spending against $71.8 billion of revenue, $46.9 billion of operating cash flow, free cash flow of negative $28.7 billion and net debt of $119.6 billion, with $20.1 billion of common stock and $5.0 billion of preferred issued in twelve months. Meta has the same shape — trailing capex of $104.4 billion against $241.6 billion of revenue and a projected $71 billion free operating cash flow deficit in 2027 — but a far larger advertising base and 26% revenue growth, which is why the two get graded differently for now. Shows what happens when the rating action arrives before the revenue does.

The next quarter is not where this resolves. What matters is the order of three events: the fourth-quarter report and the first 2027 capex guide, the launch of the Class C share class that unlocks the $60 billion program, and the forward-sale settlement. A 2027 guide materially below S&P's more-than-$225 billion assumption would be the first evidence that the rating action is changing behaviour rather than just the cost of money. Makes the 2027 capex guide the number that decides whether the outlook sticks.

The trade this rating action creates is not long or short AI in the abstract. It is long the assets that get paid on volume and short the credit that has to be refinanced. Memory, power, custom silicon and project-finance origination all sit upstream of the constraint; the compute lessors and the corporate balance sheet sit downstream of it, where a 1.0x trigger converts a growth plan into a funding problem. Makes the credit, not the demand, the tradeable variable.