Oracle’s rating falls from Watchlist to Avoid. The trigger is not the share price, which at $144.39 on August 5 sits within 3% of where July’s review found it. It is that the indicators that review nominated as decisive have since moved against the stock.

That report named four developments that would push Oracle toward Avoid and four that would push it toward Buy. Two of the first group have fired. Oracle’s five-year credit default swap spread has widened to roughly 203 basis points, its highest since 2008, after S&P Global cut the issuer rating to BBB-, one notch above speculative grade, with Moody’s holding Baa2 on a negative outlook. And the shares broke the $134.57 fifty-two-week low that July identified as its invalidation level, closing beneath it for thirteen straight sessions and bottoming at $114.99 on July 24.

None of the four Buy triggers has occurred. The most important was evidence that free cash flow is inflecting. It is moving the other way: consensus now models fiscal 2027 free cash flow at negative $47.73 billion, against negative $24.54 billion in fiscal 2026. Across two years that is roughly $72 billion of cash consumed, against $31.89 billion on hand and $167.43 billion of debt already outstanding at 5.6 times EBITDA.

The demand story remains intact, and this downgrade does not dispute it. The $638 billion backlog is contracted, fiscal 2027 revenue is forecast to grow 32.6%, and the Google Cloud agreement of July 30, embedding Gemini models into Fusion Applications, AI Agent Studio and NetSuite, is the most encouraging thing Oracle has produced in months. It attaches high-margin AI revenue to the installed base without further capital spending, and it drove a 9.2% single-day gain on August 3.

What it does not do is close the funding gap, or dilute the concentration of that backlog in a small number of unrated counterparties. Nor does it explain the distance between revenue growth of 32.6% and adjusted earnings growth of 5.5%, which is gross margin falling from 65.8% to a forecast 60.2%.

At 17.9 times forward earnings for 5.5% growth, describing Oracle as statistically cheap no longer holds. A probability-weighted fair value of $146 leaves almost no expected return in exchange for a one-in-three chance of a 45% loss.

A retest of $115 to $125, a stabilising credit trajectory, or reported rather than guided cash-flow improvement would each restore Watchlist. Until one arrives, Oracle’s equity market and its credit market cannot both be right.

Full workings appear in the complete Oracle Corporation (ORCL) report on our Reports page.