Capital Structure: Leverage, Debt Issuance, and Maturities
To fund a $220 billion capital expenditure program without depleting its massive on-hand liquidity ($61.45 billion in cash and equivalents as of Q2 2026), Amazon executed a historic debt issuance spree throughout 2026 (Minichart SG).
The 2026 Borrowing Spree
Between early and mid-2026, Amazon flooded the global bond markets, raising an aggregate sum surpassing $72 billion. This staggering sum effectively pre-funds the company's multi-year AI infrastructure rollout while locking in capital costs ahead of potential macroeconomic volatility.
The sequence of this debt accumulation unfolded as follows: 1. March 2026: Raised $37 billion in an 11-tranche U.S. dollar bond sale, which attracted $126 billion in investor orders, underscoring voracious institutional demand for Amazon credit (MLQ.ai). Concurrently, the company raised €14.5 billion (approx. $16.8 billion) in European markets (DataCenterDynamics). 2. June 2026: Raised C$13.96 billion (approx. $10 billion) in the Canadian debt markets across multiple tranches maturing between 2029 and 2056 (GuruFocus). 3. July 2026: Executed a massive 8-part U.S. investment-grade bond sale targeting $25 billion, bringing the year's total past $72 billion. Following this sale, Amazon explicitly notified underwriters that it would halt any further debt issuance for the remainder of 2026, capping the borrowing spree (Briefs.co).
Anatomy of the July 2026 $25 Billion Debt Offering
To understand Amazon's leverage profile, we must examine the pricing and duration of its most recent major $25 billion debt stack, settled on July 9, 2026.
| Tranche / Maturity | Principal Value | Interest Rate / Spread | | :--- | :--- | :--- | | Floating Rate Notes due 2029 | $750 million | SOFR Equivalent | | Fixed Rate Notes due 2029 (3-Year) | $3.50 billion | 4.600% (65 bps over Treasuries) | | Fixed Rate Notes due 2031 (5-Year) | $4.25 billion | 4.800% (80 bps over Treasuries) | | Fixed Rate Notes due 2033 (7-Year) | $3.00 billion | 5.000% (90 bps over Treasuries) | | Fixed Rate Notes due 2036 (10-Year) | $4.50 billion | 5.300% (100 bps over Treasuries) | | Fixed Rate Notes due 2046 (20-Year) | $2.75 billion | 6.000% (120 bps over Treasuries) | | Fixed Rate Notes due 2056 (30-Year) | $4.00 billion | Fixed rate (130 bps over Treasuries) | | Fixed Rate Notes due 2066 (40-Year) | $2.25 billion | Fixed rate (145 bps over Treasuries) |
Synthesis of Maturity Profile: The July 2026 issuance, filed under an Underwriting Agreement with institutions like Goldman Sachs, J.P. Morgan, and Morgan Stanley, perfectly illustrates Amazon's sophisticated liability management (Minichart SG). By stretching maturities out as far as 2066 (40 years), Amazon prevents dangerous near-term maturity walls. The exceptionally tight credit spreads—such as paying only 100 basis points over U.S. Treasuries for 10-year money—demonstrate that bond markets view Amazon's credit risk as practically sovereign-adjacent. Amazon currently carries elite investment-grade credit ratings: A1 from Moody's, AA from S&P, and AA- from Fitch (MLQ.ai).
Historical Debt Maturities and Refinancing Risk
Prior to the massive 2026 issuance, Amazon's total scheduled debt maturities stood at $68.8 billion as of its 2025 10-K SEC filing, against a total balance sheet long-term debt of approximately $133 billion (DebtCanary). With near-term maturities highly manageable (e.g., only $2.8 billion due in the immediate near-term window prior to 2026), the company's refinancing risk is essentially zero. The absolute size of the debt is immense, but relative to cash generation, it is highly structured and entirely sustainable.
Solvency and Interest Coverage
While absolute debt levels have soared past the $200 billion mark globally (when factoring in 2026 issuances on top of the 2025 legacy debt), equity analysts must look to the Interest Coverage Ratio (ICR) to determine true solvency.
Context: The Interest Coverage Ratio is calculated by dividing Earnings Before Interest and Taxes (EBIT) by the Interest Expense. An ICR below 1.0x means a company cannot service its debt from operating profits; a ratio of 2.5x is considered a warning line.
| Fiscal Quarter End | TTM EBIT ($ Billions) | TTM Interest Expense ($ Billions) | Interest Coverage Ratio | | :--- | :--- | :--- | :--- | | Q4 2022 (Low Point) | $13.34 | $2.36 | 5.6x | | Q4 2024 | $68.59 | $2.40 | 28.5x | | Q4 2025 | $79.97 | $2.27 | 35.2x | | Q2 2026 (Recent) | $93.71 | $3.33 | 28.1x (TTM) / 20.9x (Quarterly) |
Synthesis of Coverage Data: During the tech recession and fulfillment over-expansion hangover in late 2022, Amazon's interest coverage ratio temporarily dipped to a concerning but manageable 5.6x (with some strict quarterly metrics momentarily showing extreme volatility) (Stock Analysis on Net). However, the aggressive cost-cutting measures instituted by CEO Andy Jassy over 2023 and 2024 resulted in a spectacular margin recovery.
By Q2 2026, Amazon generated $93.71 billion in TTM EBIT against only $3.33 billion in TTM interest expense, yielding a coverage ratio of 28.1x for the trailing twelve months, and roughly 20.9x for the specific second quarter metrics (Finbox).
In Q2 2026 specifically, the company reported $27.5 billion in operating income against just $1.3 billion in interest expense for that three-month period (SEC.gov). In an industry group of 850 cyclical retail and tech peers, Amazon's coverage ratio ranks better than 77.6% of the market (GuruFocus). The conclusion is absolute: Amazon is heavily leveraged in absolute dollar terms, but highly under-leveraged relative to its colossal earnings power.
Valuation: Multiples and Implied Growth
Valuing Amazon has historically required suspending traditional value-investing dogmas. For much of the 2010s, the company traded at Price-to-Earnings (P/E) ratios exceeding 300x, peaking near 977x in late 2015 as the company intentionally operated at break-even to seize market share (Stockrow).
However, as of October 2026, Amazon's valuation profile has matured significantly, presenting a compelling case for Growth-at-a-Reasonable-Price (GARP) investors.
Current and Forward Valuation Metrics
Trailing P/E Ratio: As of mid-October 2026, with the stock trading around $250 - $260, Amazon sports a Trailing Twelve Month (TTM) P/E ratio of approximately 19.7x to 20.1x (GuruFocus). This represents a dramatic compression from historical norms and sits below the internet retail industry median of 23.2x. Forward P/E Ratio (Near Term): Based on consensus estimates for the next twelve months, the forward P/E ratio hovers around 31x (Motley Fool). While this implies a premium to the broader S&P 500, it remains highly attractive relative to Amazon's historical trading range (which routinely exceeded 50x forward earnings). Morgan Stanley's 2028 Forward P/E: The crux of Brian Nowak's $350 price target is grounded in long-term earnings power. Morgan Stanley projects Amazon will generate GAAP Earnings Per Share (EPS) of $16.56 by 2028. Applying a target multiple to this figure, the $350 price target implies a highly reasonable forward P/E of just 21x on 2028 earnings (Moomoo). PEG Ratio: The Price/Earnings-to-Growth (PEG) ratio stands at a healthy 1.46, sitting comfortably within the normal range for comparable mega-cap tech stocks (averaging 1.77) (Tickeron). A PEG near 1.0 to 1.5 indicates that the market is accurately pricing the equity relative to its expected EPS growth rate. Price-to-Sales (P/S): The P/S ratio sits at 3.44, slightly elevated compared to the retail industry average, but entirely justified given that the high-margin AWS segment comprises a disproportionate amount of the underlying value.
Strategic M&A and Infrastructure: Expanding the Moat
Amazon is aggressively deploying its capital not just into internal Capex, but into highly strategic acquisitions designed to insulate its future competitive positioning against both AI software rivals and aerospace disruptors.
The Decart Acquisition: Bridging the Chip Divide
In October 2026, it was revealed that Amazon is in advanced talks to acquire the Israeli-founded artificial intelligence startup Decart for approximately $7 billion (WSJ via BizJournals).
Strategic Rationale: Founded in 2023, Decart specializes in software that allows AI models to run efficiently across multiple chip types. Currently, the AI industry is heavily locked into Nvidia's proprietary CUDA (Compute Unified Device Architecture) software ecosystem. Amazon has been desperately developing its own custom AI silicon (Trainium and Inferentia) to reduce its crippling dependency on Nvidia. Decart's technology acts as a bridge, making it frictionlessly easy for developers to transition their workloads from Nvidia GPUs to Amazon's native Trainium chips. The urgency of this transition is underscored by the impressive hardware metrics of Amazon's proprietary silicon: Amazon's Trainium 3 chip, built on a 3-nanometer process, delivers 4x faster performance and 40% better energy efficiency than previous generations, directly challenging Nvidia's dominance [cite: 12]. Additionally, AWS's Inferentia2 (Inf2) instances offer 25% to 40% lower cost per inference compared to equivalent Nvidia GPU instances for supported workloads [cite: 13]. This internal silicon advantage has already secured multi-year, multi-gigawatt compute commitments from AI leaders like Anthropic and OpenAI [cite: 7].
Furthermore, Decart specializes in "world models"—AI architectures capable of simulating 3D real-world environments. If finalized, this $7 billion acquisition would be one of Amazon's largest, rivaling its $8.45 billion purchase of MGM Studios and its $13.7 billion acquisition of Whole Foods. It represents a direct, aggressive strike at Nvidia's software moat (PYMNTS).
Globalstar: The LEO Satellite Offensive
Earlier in 2026, Amazon announced a definitive $10.8 billion (with some sources citing up to $11.5 billion in total transaction costs) agreement to acquire Globalstar, a legacy mobile satellite provider (Broadband Breakfast).
Strategic Rationale: Amazon is currently racing to build 'Project Kuiper,' a Low Earth Orbit (LEO) satellite network designed to rival Elon Musk's SpaceX Starlink. Amazon faces a stringent regulatory mandate from the FCC requiring half of its planned 3,200-satellite constellation to be operational by mid-2026. Acquiring Globalstar provides Amazon with immediate access to crucial radio spectrum rights, two dozen existing operational satellites, and an established direct-to-device infrastructure (Globalstar notably powers Apple's Emergency SOS iPhone features, a contract Amazon has agreed to honor). This acquisition instantly validates Amazon as a tier-one aerospace telecommunications provider.
The Nuclear Catalyst: Powering the AI Boom
A tertiary, yet highly critical development occurred concurrently with the Morgan Stanley upgrade: Amazon's formalized entry into nuclear power generation. In October 2026, Amazon signed a funding agreement with X-energy and American Electric Power (AEP) to develop a 960-megawatt Small Modular Reactor (SMR) project in Sullivan County, Indiana (Finance BigGo).
Strategic Rationale: The 2 to 3 million Nvidia GPUs Morgan Stanley projects AWS will deploy by 2028 require an astronomical amount of uninterrupted, baseload electricity. The U.S. power grid is currently ill-equipped to handle the gigawatt-scale demands of modern AI data centers. By directly funding feasibility and development of a 12-unit Xe-100 nuclear plant on a brownfield coal site in Indiana, Amazon is vertically integrating its most critical input cost: electricity.
Risks, Red Flags, and Open Questions
Despite the euphoric reaction to the Morgan Stanley upgrade and the pristine operational metrics, senior equity analysts must scrutinize two massive, looming threats that possess the capability to derail Amazon's equity trajectory.
1. The FTC Antitrust Trial (The Existential Regulatory Threat)
The single largest exogenous risk facing Amazon is the landmark antitrust lawsuit brought forth by the Federal Trade Commission (FTC) and 17 state attorneys general. Filed in 2023 under the direction of FTC Chair Lina Khan, the case accuses Amazon of operating an illegal monopoly and utilizing anticompetitive strategies to maintain its dominance over online commerce.
The timeline is critical: A federal judge in Washington set the official trial date for October 2026 (Inc.). Because the trial is occurring presently (as of the publication of this report), the stock carries immense "headline risk."
The Core Arguments: The FTC is fundamentally attacking Amazon's fulfillment requirements (Fulfillment by Amazon, or FBA) and its pricing parity rules. The government argues that Amazon forces third-party sellers to use its expensive logistics network to obtain the coveted "Prime" badge, while simultaneously penalizing sellers who offer their goods at lower prices on competing websites.
The Red Flag: The outcome of this trial will have wide-ranging reverberations throughout the technology and M&A sectors (Axios). If the government prevails, behavioral remedies could force Amazon to decouple its Prime badge from its logistics network, effectively destroying the flywheel that has made its e-commerce operations so dominant. In a worst-case scenario, structural breakup remedies could be pursued, forcing the spin-off of AWS from the retail business. While some activist investors argue an independent AWS would command a higher multiple, the forced dis-integration of the company would trigger years of legal and operational chaos.
2. The $220 Billion CapEx Question
The second major open question revolves around the terminal Return on Invested Capital (ROIC) of the current AI infrastructure build-out.
Amazon is projecting $220 billion in capital expenditures in 2026 alone. To put this in perspective, that is roughly equivalent to the entire gross domestic product of New Zealand, spent in a single year on servers, data centers, and fiber optics.
While AWS revenue is currently re-accelerating (hitting 37% YoY growth in recent quarters), the ultimate monetization of AI remains unproven across the broader enterprise software landscape. If corporate adoption of AI agents and LLM (Large Language Model) inference plateaus, or if the productivity gains do not justify the software licensing costs for end-users, Amazon will be left with hundreds of billions of dollars in rapidly depreciating semiconductor hardware.
The $72 billion in newly minted debt ensures they have the cash to survive a build-out cycle, but the depreciation expense from these CapEx additions will create a massive headwind to GAAP net income in 2027 and 2028. Morgan Stanley's $16.50 EPS target for 2028 intrinsically assumes that AI infrastructure depreciation will be vastly outpaced by high-margin software inference revenue (which currently targets gross margins of roughly 50% to 60%, a noticeable compression from traditional SaaS margins of 80% to 90%, but still highly lucrative at scale) [cite: 14, 15]. If this hypothesis fails, the $350 price target will aggressively compress.
Conclusion
Amazon (AMZN) in late 2026 operates in a league entirely its own. It is simultaneously executing the largest capital expenditure program in corporate history, navigating a historic federal antitrust trial, and radically expanding its cloud capabilities through nuclear energy partnerships and targeted AI M&A.
The Morgan Stanley upgrade to a $350 price target correctly identifies the under-appreciated margin potential of AWS as it scales into the "Rule of 90." The company's complete lack of a dividend is a testament to its supreme confidence in internal capital reinvestment—a confidence backed by $161 billion in operating cash flow and a bulletproof 20x interest coverage ratio. However, investors entering at these levels must remain acutely vigilant. The October 2026 FTC trial poses an existential threat to the retail business model, and the race to build AI infrastructure at all costs leaves little room for macroeconomic error.
Sources: 1. aninews.in 2. tomshardware.com 3. thestreet.com 4. nextmsc.com 5. uncoveralpha.com 6. marketsandmarkets.com 7. zacks.com 8. cxtoday.com 9. amazon.com 10. cloudzero.com 11. omidsaffari.com 12. medium.com 13. wring.co 14. cloudzero.com 15. digitalapplied.com
