We Value Your Feedback! Help us shape better content and experiences by participating in our survey. Join the Survey Now!
In-Depth Equity Analysis:Oracle (ORCL)
Executive Summary
Event Recap: Oracle (ORCL) shares recently suffered a sharp decline (dropping over 11% in a single day), effectively erasing all gains made since September. The primary catalyst was the Fiscal Year 2026 Second Quarter (26Q2) earnings missing expectations. Although demand for Oracle Cloud Infrastructure (OCI) remains robust, lagging data center construction progress (delays) has pushed back revenue recognition.
Core Conflict: The market's focus has fundamentally shifted. Investors are no longer cheering solely for record-breaking Remaining Performance Obligations (RPO, reaching $523 billion). Instead, anxiety is mounting over the widening gap between soaring Capital Expenditures (Capex, raised to $50 billion) and the short-term realization of revenue. Oracle is facing cash flow pressure under a "Sell Balance Sheet as a Service" model.
Key Conclusion: This sell-off does not falsify the AI demand thesis; rather, it highlights physical supply-side bottlenecks (power, construction delivery) limiting growth speed. Oracle is in a painful transition period from a software giant to a compute infrastructure titan. In the short term, construction delays and free cash flow deficits will suppress valuation; in the long term, if delivery bottlenecks can be resolved, its unique position in the AI compute rental market remains solid.
1 Introduction — Misaligned Expectations Behind the Plunge
1.1 Market Phenomenon: A "Cold Shower" Amidst the AI Fever
Against the backdrop of the global Artificial Intelligence (AI) wave pushing tech giants to record highs, Oracle's recent correction of over 10% stands out starkly. This drop has wiped out all stock price appreciation accumulated since September. As a core ally of NVIDIA and a significant player in the AI compute rental market, Oracle had previously enjoyed a high valuation premium. However, following the 26Q2 earnings release, market sentiment reversed rapidly. This was not just an earnings "miss"; it was a market repricing of the difficulty of building AI infrastructure and the timeline for monetization.
1.2 Core Conflict: Demand is No Longer the Issue; Supply is the Bottleneck
Over the past two years, the core market logic was "He who has the cards (GPUs) is King." However, Oracle's case reveals a new phase of contradiction: Even if you have the cards, you cannot necessarily turn them into cash immediately.
The End of the Order (RPO) Euphoria: Oracle holds massive orders from giants like Meta and OpenAI, and RPO is still growing sequentially. However, this quarter, the market showed a rare desensitization to this "paper wealth."
The Beginning of Delivery (Revenue) Anxiety: Investors have realized that Oracle's performance is no longer constrained by the capability of its sales team, but by the progress of its construction crews. Delays in data center construction—specifically bottlenecks in power, cooling, and powered shell delivery—have directly resulted in the inability to recognize revenue.
1.3 Thesis of This Report
Is this crash the beginning of a deterioration in Oracle's fundamentals, or a healthy correction within a long-term uptrend?
This report will deeply analyze how Oracle is aggressively transforming from a traditional database software company into a "Compute Landlord." We will focus on:
The real impact of data center construction delays and the shock to cash flow.
How Oracle's core business model has undergone a qualitative change in the AI era.
How to assess its investment value and risks given the surge in Capital Expenditures (Capex) while revenue guidance remains unchanged.
2 Deep Dive — Delivery Crisis and Cash Flow Black Hole
2.1 Revenue and Delivery: Bottlenecked by the Physical World
The core contradiction of the 26Q2 earnings report lies not in demand, but in the physical bottlenecks on the supply side.
Revenue: Recorded at $16.1 billion, missing the expectation of $16.2 billion.
OCI (Cloud Infrastructure) Revenue: Recorded at $4.08 billion, missing the expectation of $4.17 billion.
Core Logic of the Miss:
Ingredients but No Pot: The revenue miss wasn't due to a lack of customers, but because data centers couldn't be built fast enough. Management explicitly identified Powered Shells (empty data center shells equipped with power, cooling, and networking) as the primary bottleneck.
Third-Party Delays: Partner Core Scientific (CRWV) delayed delivery on multiple projects in Texas, Oklahoma, and North Carolina. This directly prevented RPO (Remaining Performance Obligations) from being timely converted into recognized Revenue.
Guidance Divergence: These delivery delays forced CRWV to lower its full-year 2025 revenue guidance. Similarly, Oracle maintained its FY2026 total revenue guidance at $67 billion (despite a massive increase in Capex) and instead pushed growth expectations to FY2027 (raising guidance by $4 billion). This is a classic phenomenon of "Revenue Deferral."
2.2 The Truth About RPO: The Market Has Desensitized to Big Numbers
Explosive Data: 26Q2 RPO reached a staggering $523 billion (Note: If this figure represents Total Contract Value, it indicates extremely long-term lock-ins; if it is standard RPO, it shows an unprecedented backlog), with a quarter-over-quarter net increase of $68 billion. This includes massive long-term contracts from Meta and NVIDIA.
Cold Market Reaction: Despite the eye-popping numbers, the market reaction was tepid. Investors have become desensitized to "orders on hand." The current logic is: RPO that cannot be converted into revenue is just a numbers game on the books. As long as delivery bottlenecks remain unresolved, a higher RPO only serves to highlight shortcomings in execution.
2.3 Capital Expenditure (Capex) and Cash Flow: A Shocking Cash Burn
This is the most jaw-dropping part of this quarter's report and the root cause of institutional panic.
Capex Surge: The FY2026 capital expenditure guidance was raised by $15 billion to a total of $50 billion. This is intended to meet the urgent demand within that net increase of $68 billion in RPO.
Free Cash Flow (FCF) Shock: The quarter recorded an FCF of -$10 billion.
For a software giant once known for high dividends and buybacks, a single-quarter net cash outflow of this magnitude is disruptive. It means Oracle's current operating cash flow is far from sufficient to cover its infrastructure investments.
Financing for Survival: To fill this hole, Oracle already issued $18 billion in bonds in September and is preparing for a new round of $38 billion in debt financing. This financial model completely mirrors that of a utility company or a real estate developer.
3 Business Model Reconstruction — From SaaS Dominance to "Balance Sheet as a Service"
To understand why Oracle dares to gamble heavily even when facing a cash flow gap of tens of billions, one must first understand its unique business landscape and the fundamental transformation occurring within its business model.
3.1 The Cornerstone of Business: Not Just Databases, But the "Digital Nervous System" of Enterprises
Oracle's business extends far beyond the databases known to the general public; it effectively controls the core lifelines of large global enterprises. Its operations are primarily divided into two main sectors:
Core Application Software (SaaS - Software as a Service):
ERP (Enterprise Resource Planning): This is Oracle's ace in the hole (Oracle Fusion Cloud ERP). It manages a company's purse strings (finance), ledgers, and daily operations.
HCM (Human Capital Management): Manages employee recruitment, payroll, and performance.
SCM (Supply Chain Management): Manages the entire process from raw material procurement to product delivery.
Customer Profile & Stickiness: Downstream customers are primarily Global Fortune 500 companies, large government agencies, and complex multinational corporations. These clients have extremely high requirements for software stability, compliance, and data security.
The Moat: Extremely high switching costs. Once a multinational bank or airline adopts Oracle's ERP system, the cost of replacing it involves not just money, but massive business interruption risks. This characteristic of "once adopted, bound for life" provides Oracle with a continuous stream of cash flow.
Infrastructure (IaaS - Infrastructure as a Service):
This is OCI (Oracle Cloud Infrastructure), the battlefield where Oracle has been striving to catch up with AWS and Azure in recent years. It provides computing power, storage, and networking—the physical foundation for running the SaaS software mentioned above and, more recently, the booming AI large models.
3.2 Evolution of the Business Model: From "Selling Software" to "Selling Compute Real Estate"
Oracle is undergoing the most radical business model switch since its inception, which we term "Balance Sheet as a Service."
The Traditional Model (High Margin, Asset-Light):
In the past, Oracle made money by selling software licenses and maintenance fees. This was a fantastic business: write the code once, sell it to ten thousand customers, marginal costs were near zero, gross margins were extremely high, and cash flow was abundant. This was the foundation that allowed Oracle to aggressively buy back stock and pay dividends over the past few decades.
The Current Model (Asset-Heavy, High Leverage):
With the advent of the AI era, customers are no longer satisfied with just buying software; they need massive computing power to train models (e.g., OpenAI, xAI, NVIDIA). Oracle identified a huge opportunity, but it comes with a huge price tag:
Core Logic: Oracle utilizes its powerful balance sheet (borrowing capacity) to issue massive amounts of debt (such as the upcoming $38 billion financing) to purchase expensive NVIDIA GPU chips and build data centers.
Subleasing to Customers: It then "rents" these expensive computing resources in the form of cloud services to AI startups or tech giants that desperately need computing power but do not want to build their own data centers.
Fundamental Change: Oracle has transformed from a pure software company into a "Digital Real Estate Developer." It borrows money to construct buildings (Data Centers + GPUs) and then collects rent (Cloud Service Fees).
3.3 Core Barriers and Competitive Advantages
Since it is acting as a "landlord," how does Oracle compete with Amazon (AWS) and Microsoft (Azure)?
Differentiated Network Architecture (RDMA): Oracle's OCI was designed from the ground up with optimization for high-performance computing. Simply put, its network allows thousands of GPUs to work collaboratively like a single super-chip with extremely high efficiency, which is exactly what training large AI models requires most.
Data Sovereignty & Hybrid Cloud Advantage: Leveraging its dominance in the traditional database field, Oracle can offer unique "Sovereign Cloud" services, allowing governments or sensitive industries to keep data entirely local—a compliance advantage other cloud vendors struggle to match.
Price Aggressor Strategy: To seize market share, Oracle is often more aggressive than AWS in data transfer fees and compute pricing, attracting a large number of cost-sensitive AI unicorns.
3.4 Characteristics and Risks of the Business Model
This new model is a double-edged sword:
Characteristics: High Upfront Investment, Long Payback Period. It requires spending $50 billion now to buy equipment (Capex), while revenue is collected slowly over the next few years through monthly customer bills.
Risk Points:
Cash Flow Black Hole: As mentioned in Part 2, free cash flow will turn negative in the short term, necessitating reliance on external financing for "blood transfusions."
Technology Depreciation Risk: GPUs update extremely fast. If the H100 chips Oracle borrows money to buy now become obsolete in two years, and customers switch to the next generation, these massive investments could turn into bad debts.
Execution Test: This model demands extremely high delivery speed. If data center construction is slow (like the power and shell delays this quarter), there is no revenue, yet the massive interest payments remain.
Summary: Oracle is using the deep customer trust and cash flow reserves accumulated in the SaaS sector to leverage its way into the capital-intensive AI computing market. This is a gamble on the durability of AI demand and its own ability to deliver as a "digital infrastructure builder."
4 Valuation and Final Verdict — A High-Stakes Poker Game
We have dissected Oracle’s concerning cash flow gaps and its ambition to transform via AI infrastructure. Now, we must scrutinize all of this through an investor's lens.
The market is currently pricing Oracle not as a traditional legacy software company, but as a high-growth AI infrastructure giant. This creates a dangerous "valuation disconnect."
4.1 Valuation Analysis: The Battle Between "AI Premium" and Financial Reality
Multiple Expansion:
Historical Norm: Historically, Oracle’s P/E Ratio (Price-to-Earnings) has hovered between 15x and 18x. This is typical for a mature, slow-growth tech giant.
Current Reality: Today, its forward P/E is often pushed to 30x or higher. This means investors are no longer valuing it based on past profits, but are paying a premium for future growth.
The Mismatch of Expectations:
The current stock price implies a hypothesis: Oracle will sustain double-digit revenue growth for years to come.
However, current financial statements show a company burning cash frantically to build capacity. Free Cash Flow (FCF) has not kept pace with the stock price rally.
The Danger Zone:
When a company possesses both High Debt + High Valuation, its margin for error is zero.
If Oracle misses an earnings expectation, or if the AI market heat cools even slightly, the stock faces a "Double Whammy":
Earnings Per Share (EPS) drops.
Valuation multiples contract (investors stop viewing it as an AI play, and P/E falls back to 15x).
This is the scenario the "Big Short" players are betting on.
4.2 The Bear Case: Why are Skeptics Shorting?
The logic of the short sellers is built on three solid pillars of risk:
Margin Dilution (The "Bad Revenue" Problem):
Oracle’s traditional business (SaaS/ERP) boasts extremely high gross margins of 90%+. That is a pure money printer.
The new business (IaaS/Compute Rental) has much lower margins (often only 30%-50% after depreciation and power costs).
The Trap: As Oracle’s AI revenue share increases, its overall profit margin must mathematically decline. They are trading high-quality, asset-light revenue for low-quality, asset-heavy revenue.
Execution Risk (Building Data Centers is Hard):
Larry Ellison has promised to build massive data centers powered by small nuclear reactors. This sounds like science fiction, but the reality is harsh.
Construction faces multiple challenges: power shortages, regulatory approvals, and Nvidia chip supply bottlenecks. Any delay turns Capital Expenditures (CapEx) into dead money.
Questionable Competitive Moat:
Oracle’s current advantage is being "fast" and "cheap." But what if Amazon (AWS) and Microsoft (Azure) decide to start a price war? What if they resolve their GPU shortages?
Oracle lacks an irreplaceable ecosystem moat like Nvidia’s CUDA or Microsoft’s Windows. Fundamentally, it is just renting out hardware.
4.3 The Bull Case: Why Larry Ellison Might Be Right
Despite the massive risks, the Bulls have strong arguments:
The Irreversible AI Wave:
If AI is indeed the Fourth Industrial Revolution, the demand for compute will be infinite. Under this assumption, Oracle doesn't need to beat Amazon; it only needs to capture a slice of this massive incremental market to make a fortune.
The current GPU supply-demand relationship is extremely unbalanced—if Oracle has the chips, they will sell.
An Extremely "Sticky" Legacy Customer Base:
While the traditional database business is slow-growing, it is incredibly hard to replace. The core data of the Global 500 resides on Oracle.
Oracle’s strategy is simple: Since your data is already here, why not use my cloud to train your AI models directly? This "Data Gravity" is its greatest invisible asset.
Ellison’s "Reality Distortion Field":
Never underestimate Larry Ellison. He is one of Silicon Valley’s shrewdest tacticians. He successfully pivoted Oracle from on-premise to SaaS, and now he is betting on IaaS. History proves he is often right at critical moments.
4.4 Final Verdict: Investment Advice
What kind of stock is Oracle?
It is no longer a stable dividend stock suitable for "widows and orphans." It is now a leveraged AI infrastructure cyclical stock.
For Aggressive Investors: If you firmly believe that AI Scaling Laws will continue to hold true and that the compute shortage will last for 3-5 years, then Oracle is one of the best "pick-and-shovel" plays outside of Nvidia. Its upside comes from successfully stealing market share from AWS.
For Conservative Investors: AVOID. The balance sheet is too weak (high debt), cash flow is too poor (eaten by CapEx), and the valuation has priced in two years of perfect execution. If the macro economy enters a recession and enterprises cut IT spending, Oracle will be one of the hardest-hit tech giants.
Summary Conclusion:
Oracle is making a massive gamble. It has mortgaged its future as a high-margin software company to transform into a heavy-asset utility company (a compute factory).
If it succeeds: It will become the "Power Grid" of the AI era, and its market cap could double.
If it fails: Massive debt and excess data centers will become a heavy anchor, causing the stock price to be cut in half.
This explains why its stock price is so volatile and why Wall Street’s view on it is so polarized. This isn't just an investment; it is a binary bet on the future of AI.
5 Risk Factors
CapEx Efficiency Risk: Oracle is currently undertaking aggressive capital expenditure (projected $10B+ annually) to build AI infrastructure. If demand for AI compute softens or fails to materialize as projected, these assets could become a drag on Return on Invested Capital (ROIC) and Free Cash Flow.
Macroeconomic Headwinds: A global economic slowdown could lead enterprise customers to reduce IT budgets. Since Oracle relies heavily on large enterprise contracts, a recession would directly impact both its legacy license revenue and cloud consumption rates.
Competition & Pricing Pressure: The Cloud Infrastructure (IaaS) market is dominated by hyperscalers (AWS, Azure, Google Cloud) with deeper pockets. If these competitors initiate a price war to protect market share, Oracle’s operating margins could be severely compressed.
Debt Management: Oracle carries a significant debt load relative to its cash position. Rising interest rates or a failure to generate sufficient cash flow could limit the company's ability to deleverage, pay dividends, or pursue strategic acquisitions.
Geopolitical & Supply Chain Constraints: The company’s AI strategy is dependent on the supply of Nvidia GPUs and the construction of data centers. Supply chain disruptions or geopolitical tensions (e.g., chip export restrictions) could delay capacity expansion and revenue recognition.
Disclaimer:
The Information presented above is for information purposes only, which shall not be intended as and does not constitute an offer to sell or solicitation for an offer to buy any securities or financial instrument or any advice or recommendation with respect to such securities or other financial instruments or investments. When making a decision about your investments, you should seek the advice of a professional financial adviser and carefully consider whether such investments are suitable for you in light of your own experience, financial position and investment objectives. The firm and its analysts do not have any material interest or conflict of interest with any stocks mentioned in this report.
IN NO EVENT SHALL SAHM CAPITAL FINANCIAL COMPANY BE LIABLE FOR ANY DAMAGES, LOSSES OR LIABILITIES INCLUDING WITHOUT LIMITATION, DIRECT OR INDIRECT, SPECIAL, INCIDENTAL, CONSEQUENTIAL DAMAGES, LOSSES OR LIABILITIES, IN CONNECTION WITH YOUR RELIANCE ON OR USE OR INABILITY TO USE THE INFORMATION PRESENTED ABOVE, EVEN IF YOU ADVISE US OF THE POSSIBILITY OF SUCH DAMAGES, LOSSES OR EXPENSES.
