Oracle is one notch above junk after S&P downgrade as AI data-centre spending burns through cash

S&P downgraded Oracle to BBB-, moving it one notch above junk, citing that its $250B data-centre expansion plan is burning cash faster than revenue can replace it. Oracle now has $117B outstanding and is the second-largest non-financial issuer in the Bloomberg US Corporate Bond Index (after Amazon). Shares fell on the downgrade amid worsening credit-risk concerns.

Analysis

The first-order damage is not the rating action itself; it is the forced repricing of Oracle’s funding model. Once a hyperscaler starts looking like a credit spread story, incremental AI/data-center capex becomes less accretive because every dollar of growth now carries a higher all-in cost of capital, which is the right lens for why the equity multiple can compress faster than the earnings model changes.

Second-order, this is a relative win for the balance-sheet winners in AI infrastructure, especially AMZN, which can keep spending aggressively without the same near-term credit overhang. The market may start distinguishing between “strategic capex” and “levered capex,” which is important because any enterprise customer worried about vendor longevity will prefer platforms backed by stronger free cash flow and lower refinancing risk.

The key risk is that the downgrade is a lagging signal: if Oracle converts the buildout into faster OCI growth within 1-2 quarters, spreads could stabilize and the equity could rebound on a “worst is known” move. What would falsify the bearish view is a clear inflection in free cash flow or a sharp tightening in ORCL bond spreads; absent that, this is more likely a months-long story than a one-day event.

Key Decisions for Investors

• Short ORCL on strength for the next 1-3 months; use the downgrade-driven rally, if any, to lean into a 10-15% downside target in the equity with the key stop being a sustained recovery in free cash flow guidance.
• Pair trade: long AMZN / short ORCL for a 6-12 week relative-value view on cost of capital and AI infrastructure execution; this is a cleaner expression than outright tech beta if credit markets remain risk-off.
• For options-sensitive exposure, buy ORCL put spreads 1-2 months out rather than naked puts; the thesis is spread widening and multiple compression, but a near-term stabilization in the bond market could cap downside.
• Set a watch item on ORCL bond spreads and any management commentary on capex pacing; if spreads tighten materially or capex is reduced, the short thesis loses its catalyst and should be cut.