Oracle: Slumping Despite Big Beat

Oracle delivered Q4 revenue growth of 21% and EPS growth of 24%, beating consensus on both metrics. Cloud infrastructure drove nearly all growth, with revenue up more than 90% year over year and remaining performance obligations reaching $638 billion. However, heavy cloud capex produced negative free cash flow of $24 billion over the last twelve months, limiting shareholder returns and forcing $40 billion of new equity and debt issuance.

Analysis

ORCL is turning into a capital-allocation story disguised as a growth story. The near-term winner is not just the equity holder but the entire AI infrastructure stack: more Oracle capex means incremental demand for networking, power, memory, and outsourced deployment capacity, while hyperscale rivals have to defend share by keeping their own build rates high. The second-order loser is ORCL’s own capital return narrative—once a company is forced to fund growth with outside capital, the market starts valuing it less like a mature software compounder and more like a utility-style infrastructure buildout with execution risk.

The key risk is that this becomes a long-duration balance sheet trade rather than a fundamentals rerating. Backlog visibility can support the stock for months, but if conversion rates slip or customer concentration shows up in working-capital swings, the market will start discounting the backlog at a lower quality multiple. The path dependency matters: the next 1-2 quarters are about proving that booked demand turns into billings and cash, while the next 12-24 months are about whether depreciation, financing costs, and ongoing capex suppress free cash flow enough to cap multiple expansion.

Contrarianly, the bullish consensus may be underestimating how much of the upside is already tied to supply, not demand. If ORCL’s cloud supply remains constrained, revenue can stay strong but margins and cash generation may lag long enough for the equity to de-rate even as headline growth looks excellent. That creates an interesting split: the best expression of the AI infrastructure trade may be owning the enablers with cleaner FCF and avoiding the names that must continuously issue capital to keep pace.

Key Decisions for Investors

• Stay long ORCL only via a defined-risk structure: buy 3-6 month call spreads instead of outright equity to capture backlog-driven momentum while limiting damage if FCF stays deeply negative and the market focuses on dilution/financing risk.
• Pair trade: long MSFT or AMZN against short ORCL over the next 1-3 months. The thesis is that the market will increasingly reward platforms with stronger self-funded capex and penalize the one needing external capital, especially if AI demand broadens beyond the current winner.
• Overweight the infrastructure enablers, not ORCL itself: long ANET/AVGO on 6-12 month horizon if you want AI capex exposure with better cash conversion. Risk/reward is cleaner because they monetize the buildout without the same balance sheet strain.
• If you want a tactical short, wait for any post-earnings/guide pop and short a partial position for 4-8 weeks. The setup improves once momentum traders have chased the backlog headline and the market refocuses on financing needs and capital returns.
• Watch ORCL credit spreads and new-issue concessions closely for the next 30-90 days; widening spreads would be an early signal that equity investors are late to the leverage story and would justify reducing exposure.