Oracle Stock Collapsed 67%: Bargain or Debt-Fueled Disaster?

The Collapse

On September 10, 2025, Oracle hit an intraday high of $345.72. By July 28, 2026, the stock had fallen to $114.50 — a peak-to-trough decline of approximately 66.9%. More than $518 billion in market capitalization was erased. The stock has since rebounded to around $138.65, roughly 21% above the low but still nearly 60% below its high. To put that drawdown in context, Oracle at its peak was approaching a $922 billion market cap. Today it sits around $404 billion.

The decline happened in stages. In December 2025 the stock fell approximately 13% in a single session after disappointing revenue and higher-than-expected capex guidance erased around $90 billion in market value in 1 day. On June 11, 2026 it fell another 12% after Oracle disclosed even larger AI infrastructure spending and financing plans, wiping out roughly $72 billion more. By late July, concerns about debt levels, credit quality, and Oracle's exposure to a small number of enormous AI customers pushed the stock to a fresh 52-week low. The recent bounce has been helped by an expanded Google Gemini partnership and strong Azure results from Microsoft, which gave investors renewed confidence that cloud demand is real and that Oracle is genuinely in the middle of it.

The Growth Is Real and It Is Massive

Before getting into the danger, the bull case deserves to be stated clearly — because the growth numbers are genuinely extraordinary. Oracle generated $67.4 billion in revenue in fiscal 2026, up 17% year over year. Cloud revenue grew 39% to $34.0 billion. OCI infrastructure revenue — Oracle's cloud computing platform — grew 77% to $18.1 billion. Oracle Multicloud AI Database revenue grew 404% in the 4th quarter alone. Operating cash flow grew 54% to $32 billion. And management is guiding fiscal 2027 revenue to approximately $90 billion, implying approximately 34% growth for a company that was already generating nearly $58 billion in annual revenue entering the year.

The backlog is equally staggering. Oracle ended fiscal 2026 with $638 billion of remaining performance obligations — essentially contracts for future revenue. That number is up 363% year over year. For reference, that backlog is larger than the GDP of most countries. The catch is that only 12% is expected to become revenue in the next 12 months, with another 34% converting over the following 2 years. Oracle must spend the capital today to build capacity for revenue that may not fully arrive for several years. That timing mismatch is at the heart of everything investors are debating right now.

The Debt and the Danger

Here is where the story gets genuinely complicated. Oracle's non-current borrowings increased from $85.3 billion to $122.3 billion during fiscal 2026 alone. Including the current portion, total notes and borrowings rose from approximately $92.6 billion to $129.5 billion — a 40% increase in 1 year. Oracle raised $43 billion through debt during fiscal 2026 and approximately $5 billion through equity. For fiscal 2027 it is planning another $40 billion in combined debt and equity financing, including a $20 billion at-the-market equity issuance that will dilute existing shareholders. The balance sheet carries approximately $31.9 billion in cash, but much of that has effectively already been earmarked for infrastructure spending, leaving net borrowings of roughly $98 billion.

On July 9, S&P Global downgraded Oracle's issuer credit rating from BBB to BBB- — 1 notch above speculative grade. S&P cited rising business risk and weakening cash flow. That is the kind of event that tends to be dismissed by equity investors in a bull market and remembered in a bear one. Equity investors are debating how much Oracle might earn. Credit investors are debating how much risk Oracle can carry. Both conversations are happening simultaneously and they are not telling the same story.

The capital expenditure picture is the most dramatic element of the whole thesis. Oracle spent $55.7 billion on capex in fiscal 2026 — approximately 83% of its entire annual revenue. For fiscal 2027, Oracle's gross infrastructure buildout could reach $90 billion to $95 billion against guided revenue of approximately $90 billion. Net of customer reimbursements of $20 billion to $25 billion, Oracle's own net capex lands around $70 billion — still roughly 78% of revenue. There is no modern precedent for a company of Oracle's size spending at this ratio. The result is that despite generating $32 billion in operating cash flow in fiscal 2026, Oracle produced approximately negative $23.7 billion in free cash flow. The cash is leaving the business far faster than it is being earned.

There is also a customer concentration risk that deserves attention. Reports surrounding the S&P downgrade estimate that OpenAI alone may represent close to half of Oracle's $638 billion contracted backlog. Oracle does not publicly break out the exact concentration, which makes the risk difficult to measure precisely. But when the financial health of 1 customer potentially determines the fate of hundreds of billions in future revenue, that is a risk that belongs in any honest analysis.

Stock Analyzer: ORCL

Using a 13% revenue growth assumption, a 22% profit margin, and a PE of 19, the fair value comes out at $193.59 against a current price of $136.44. The implied current price return at the is 13.59% annually.

That fair value of $193.59 assumes Oracle's margins normalize to approximately 22% on a significantly larger revenue base — a reasonable assumption if the capex cycle peaks and the backlog converts into cash as management expects. But it is critical to understand what that assumption requires: that $90 billion in fiscal 2027 revenue materializes, that OCI margins improve as utilization rises, that major customers honor their contracts, that the $20 billion equity issuance dilution is absorbed without destroying per-share value, and that $150 billion-plus in cumulative infrastructure spending eventually earns returns above Oracle's cost of capital. Every one of those conditions is plausible. None of them is guaranteed.

Oracle does not need its AI strategy to fail for shareholders to be disappointed. The company could deliver enormous revenue growth and still destroy value if the returns on that capital fail to exceed the cost of building it. That is the most important sentence in this entire analysis. Tag me in the community with your own assumptions on the margin recovery. That is the number that determines everything here.

Disclaimer
Everything Money is Not an Investment Advisor: Everything Money (including Paul, Mo, and Any other person including, but not limited to, other staff members, guests, personalities, etc.) is not an investment adviser, and it is not registered as such with the U.S. Securities & Exchange Commission or any other state or federal authority under the Investment Advisers Act of 1940 or any other law. The investments and strategies discussed in Everything Money’s YouTube videos and on Everythingmoney.com are not and should not be considered investment advice and may not be suitable for you. They do not take into account your particular investment objectives, financial situation, needs, or personal circumstances and are not intended to be specific to you. Before acting on any investment or strategy discussed, you should always do your own research and make your own independent decision about whether it is suitable for your particular circumstances. You should also consider seeking advice from your own legal, financial, tax, accounting, or investment advisers. Everything Money does not provide such advice.

READ THE FULL DISCLAIMER HERE: https://everythingmoney.com/disclaimer