Oracle's Stock Crash Has Cost Larry Ellison $213 Billion in 10 Months. Should Investors Buy the Dip?

Oracle’s shares are down more than 50% from peak, wiping about $213B from Larry Ellison’s fortune, as investors question the return profile of aggressive AI data-center spending. Capital expenditures rose to over $21B in fiscal 2026 (from ~$7B a year earlier) and are expected to exceed $25B in fiscal 2027, while S&P Global Ratings downgraded Oracle to BBB- (one notch above junk), which raises financing costs for its capital-intensive expansion. Despite strong demand and a record $638B of remaining performance obligations (as of May 31, fiscal 2026), the article argues sentiment has deteriorated and risks remain if enterprise AI adoption slows.

Analysis

The market is no longer underwriting ORCL as an AI growth story; it is treating it as a leverage-and-funding story. That matters because once capex is financed at a higher spread, incremental cloud revenue has to clear a much higher hurdle rate before it supports equity value, which tends to compress the multiple long before the P&L shows stress. The immediate loser is ORCL equity; the second-order loser is any AI build-out story that depends on a benign credit market, because investors will now scrutinize whether growth is being bought with debt rather than earned with utilization.

The main beneficiaries are the better-capitalized platform names, especially MSFT, AMZN, and GOOGL, if enterprise customers decide they want AI capacity without balance-sheet risk. Oracle’s partner model also means the pain is less about losing demand outright and more about timing: if financing tightens, projects get stretched, which can create a 1-3 quarter air pocket for adjacent hardware, networking, and colocation demand even if the secular AI theme survives. NVDA should not be faded on this headline alone, but it becomes more vulnerable if delayed deployments turn into order pushouts rather than simple backlog conversion.

Contrarian view: consensus may be over-penalizing near-term ROIC risk and underweighting the scarcity premium for capacity in a supply-constrained market. The thesis breaks if ORCL can show that contracted workloads convert into cash flow fast enough to offset higher funding costs, or if management trims FY27 capex without sacrificing backlog growth. Watch credit spreads and the next capex/FCF update: if BBB- paper stabilizes and capex is revised down, the selloff can reverse quickly; if not, this is a months-long multiple problem, not a days-long bounce story.

Key Decisions for Investors

• Short ORCL on rallies or use 1-3 month put spreads; the trade is about multiple compression from financing risk, not a collapse in demand. Falsify on any meaningful reduction in FY27 capex or evidence that OCI margins are inflecting faster than expected.
• Pair trade: long MSFT or AMZN vs short ORCL over 3-6 months. Favor the names with lower funding risk and broader enterprise wallet share if AI spend stays robust but capital gets rationed.
• Do not chase NVDA purely on the Oracle headline; treat it as a watch item for delayed deployments. Reassess only if cloud capex commentary starts pointing to order pushouts rather than normal timing noise.
• Set a credit-market alert on ORCL bonds/spreads. If the downgrade bleeds into wider spreads over the next 4-8 weeks, add to the short; if spreads stabilize quickly, cover part of the position because the market has already discounted a lot of bad news.