Oracle gave Larry Ellison and his co-CEOs nearly $1 billion in stock options. By fiscal year end, all were underwater

Oracle shares closed at $137, down 53% over 12 months, leaving $988 million of fiscal-2026 option grants to Larry Ellison and co-CEOs Clay Magouyrk and Mike Sicilia underwater; their strike prices range from $280 to $308. The decline comes despite cloud revenue rising 39% to $34 billion, cloud infrastructure revenue climbing 77% to $18.1 billion, and remaining performance obligations surging to $638 billion from $138 billion. Investors are focused on the financing burden from $55.7 billion in capital spending, negative free cash flow of $23.7 billion, and $43 billion of senior-note issuance, while Oracle has also conducted layoffs and its debt-default insurance costs reportedly reached a record.

Analysis

The investable issue is not executive-option optics; it is the widening gap between contracted demand and the balance-sheet capacity needed to deliver it. ORCL is effectively underwriting a multi-year infrastructure build with negative free cash flow and materially higher fixed financing costs, so each incremental delay in data-center commissioning pushes cash conversion further out while depreciation and interest expense begin immediately. The market is likely discounting a lower-quality earnings profile: revenue growth can remain strong while equity value falls if capex intensity, lease obligations, and credit spreads continue to reset higher.

Underwater, high-strike leadership awards create an unfavorable incentive asymmetry over the next 12-24 months. Management has little near-term equity compensation value to protect, but a strong reason to pursue high-beta capacity expansion, aggressive customer commitments, or eventually seek retention/repricing relief; the November pay vote is a governance catalyst rather than a fundamental one. Broader employee option dilution also raises execution risk in scarce cloud-engineering and data-center operations roles, benefiting better-capitalized hyperscalers such as MSFT, AMZN, and GOOGL in talent retention and customer confidence.

Consensus may be too focused on the headline backlog as a valuation floor. The relevant question is conversion quality: whether booked commitments produce revenue and operating cash flow before further equity or debt issuance is required. A sustained narrowing of capex-to-revenue, stabilization in credit-default-swap levels, and evidence that cloud gross margin is expanding despite depreciation would falsify the bearish balance-sheet thesis; absent that, a rerating requires more than backlog growth.

Key Decisions for Investors

• Maintain an underweight/short bias in ORCL over the next 3-6 months, sized against a long MSFT or AMZN hedge. The pair isolates ORCL's financing and execution risk from broad AI-infrastructure demand; reassess if ORCL demonstrates two consecutive quarters of improving free-cash-flow conversion and no incremental external financing.
• Use the Nov. 18 shareholder vote and the next earnings/capex update as catalyst windows to add to an ORCL short only if management defends elevated build spending without a credible timetable for free-cash-flow normalization. A close above $185 accompanied by lower capex guidance and tightening credit spreads would invalidate the near-term short thesis.
• Avoid treating ORCL's contracted-demand disclosures as equivalent to near-term earnings visibility until customer concentration, cancellation protections, deployment milestones, and associated financing obligations are independently quantified. This is a watch item for a potential long only after backlog converts into cash rather than additional capital commitments.
• For AI infrastructure exposure, prefer diversified beneficiaries with stronger balance sheets over single-customer equipment stories tied to ORCL's build cycle. Monitor VRT and ETN for order commentary: any Oracle-related project deferral would be an early second-order signal of tighter infrastructure financing, though not yet sufficient for a standalone short.