According to Zhitong Finance APP, Apple (AAPL.US) reported strong results for its fiscal Q4 2026—revenue, EPS, and iPhone sales all exceeded expectations and hit record highs—yet its stock declined after hours. What the market is truly pricing in is not this “stellar” earnings report, but three underlying fractures hidden behind the numbers: two consecutive quarters of negative expectation gaps in Greater China and services revenue, weakening gross margin guidance amid rising storage costs, and an upcoming December quarter that has yet to unfold. At a historic high P/E ratio of 36, this defensive asset—whose $5 trillion market cap was propelled by risk-off capital seeking refuge in its avoidance of AI spending wars—is now being eroded from the cost side by the very same AI wave. For the first time, its safety requires revalidation.
Apple reported its fiscal fourth quarter 2026 results ended September 26: total revenue of $114.477 billion, a 12% year-over-year increase, exceeding LSEG estimates.
The analyst consensus of $113.39 billion also exceeded the upper end of the company’s prior guidance range of 9% to 11%; diluted EPS came in at $2.10, a 14% year-over-year increase, above expectations of
$1.99, exceeding expectations even after excluding approximately $0.05 per share from tariff refund contributions; iPhone revenue reached $57.583 billion, up 17% year-over-year, setting a new record for September
Historical quarterly financial records. Annual revenue of $479 billion, referred to by the company as "the best fiscal year ever."
However, after the earnings announcement, Apple’s stock initially rose about 0.5% in after-hours trading, then reversed and fell approximately 1%, with the decline widening further during the earnings call—by the time of this report, it was down about 3%. This isn’t an earnings disaster; it’s a case of an expectations-based trade and a stress test of the safe-haven narrative.
The top line is stable, with cracks elsewhere in three locations.
Top-line figures largely exceeded expectations. Excluding iPhone, Mac revenue reached $10.792 billion, a 24% year-over-year increase, significantly surpassing the expected range of $9.6 billion to $9.7 billion; revenue in the Americas
$48.68 billion (expected $48.51 billion), Europe $32.01 billion (expected $30.21 billion), both better than expected. Gross margin of 47.9%, above the expected 47.5%.
Cracks appeared in the other three items.
Greater China revenue was $17.539 billion, up 21% year-over-year, but $1 billion below market expectations. Service revenue was $31.573 billion, up 9.8% year-over-year, on an LSEG basis
The figure is roughly in line with expectations (31.58 billion), but this growth rate has hit the lowest level since 2022—it’s not “below expectations,” but rather a “step-down in growth pace.”
Gross margin guidance for the December quarter is 45% to 46%. The company explicitly stated that this forecast incorporates the impact of storage chip costs and Siri AI costs, while projecting revenue growth of 7% to
10%, with service revenue growth guidance of 8% to 9%, noting that third-party licensing and exchange rates pose headwinds. The revenue guidance itself exceeds the Wall Street average expectation.
6.1%, but the gross margin guidance fell below market-implied expectations.
Individually, these three flaws are not fatal, but when they appear together on a report that has already priced "security" at its maximum, their weight is amplified.
For two consecutive quarters, the same group of sectors created an expectation gap.
What truly alerts the market is not this quarter alone, but the fact that its shape is nearly identical to that of the previous quarter.
In the third quarter of fiscal year 2026, released in July, revenue in Greater China amounted to $18.82 billion, below the expected range of approximately $19.5 to $19.6 billion, with service revenue at $30.74 billion.
billion, below the expected range of approximately $31.2 to $31.4 billion. A quarter later, the same two sectors again generated negative expectation gaps simultaneously.
A one-quarter earnings surprise is a flaw; two consecutive quarters of earnings surprises in the same sector become a pattern.
The pressure on both segments is structural. Greater China faces competition from local premium brands: third-party statistics show that Apple's share in the above-$600 price segment in China has declined from approximately 75% in 2022
Reverted to approximately 50%; compounded by limited service localization and delayed AI feature rollout in the Chinese market, these pressures cannot be resolved in a single quarter.
The service, however, follows a different curve. Slowing growth in the App Store and regulatory changes have prevented Apple from collecting a 15% to 30% fee on transactions outside the App Store.
Commissions and Google search authorization revenues have slowed—these are ongoing, not one-time, headwinds. The significance of services lies in their role as Apple’s profit buffer: when hardware gross margins are squeezed by storage costs, the high margins of services should offset them; but as service growth slows to the low single digits, this buffer is thinning.
The timing when the buffer stock will be depleted is predictable. Cook previously stated that the company’s pre-purchased low-cost DRAM/NAND will gradually expire starting in June, and the inventory will be exhausted; CFO Kevin O’Leary in July
On the monthly earnings call, it was stated that memory costs accounted for more than 100% of the sequential change in adjusted gross margin. The September quarter was when the protective layer began to peel away, and the December quarter—during which gross margin guidance was lowered to 45% to 46%—
This quarter—marked by intensified erosion: the iPhone 18 Pro contributed only about nine days of sales during the September quarter, with price increase benefits not yet fully realized, while higher storage costs have already entered the cost of goods sold.
The Safe Harbor Paradox: Defense Comes from "What Was Not Done"
Apple's relative strength this year is essentially due to others spending heavily. Over the past four reported fiscal quarters, Microsoft, Amazon, Google-A, and Meta have collectively invested over $600 billion in capital expenditures, the vast majority directed toward AI data centers; Apple’s capital expenditures during the same period amounted to only about $10 billion.
Billions of dollars, less than one-sixtieth of the total of the four combined. The cost has already appeared on the financial statements of competitors: Alphabet, Amazon, Meta.
Recent disclosed free cash flows have all been negative, forcing a greater shift toward debt markets, while interest rates are rising. Apple’s free cash flow for fiscal year 2026 is expected to increase by 38% to $142 billion.
The choice of funds thus becomes mechanical: In the 46 trading days when the Philadelphia Semiconductor Index fell 2% or more, Apple rose on 30 days, accounting for approximately two-thirds; while in 1994
From last year to this year, Apple rose on only about one-sixth of these trading days. Apple’s correlation with the semiconductor index turned negative in July and reached its historical low in September. On October 30, Apple closed at $5.03
A market capitalization of $1 trillion surpassing NVIDIA's $4.93 trillion.
The issue is that this defensive attribute comes from "what Apple did not do," rather than "what Apple achieved."
More sharply, AI
The expansion of computing power is inversely impacting Apple’s profit statement through cost inputs. Calculations show that two years ago, Apple’s flash memory purchases were nearly twice the combined total of Microsoft, Amazon, Meta, and Google; today, these four companies’ combined purchases are twice that of Apple. Apple’s flash memory spending this year is projected to be approximately
$25 billion, whereas purchasing the same amount of storage in 2025 would cost only about $6 billion. According to TechInsights data, AI over the past year
Data center demand has driven flash prices up to five times their original levels. Cook compared the memory price surge to a "hundred-year flood," and, according to insiders, Apple's leadership was unprepared for this development.
In this shift of bargaining power, Apple has, for the first time in twenty years, found itself on the pressured side. This means Apple’s defense includes two opposing exposures: it does not participate in the capital expenditure race, thus avoiding scrutiny over cash flow; but since it purchases storage, it directly bears
The surge in AI computing power has driven up component prices. The risk-averse capital flowing into it is tactical and reversible—once the semiconductor sector rebounds, there’s no reason for this money to remain in Apple.
36x P/E: Zero Tolerance for "Safe Expectations" Under High Valuation
The direct context for the post-market decline is valuation. Ahead of this earnings report, Apple’s expected P/E ratio was approximately 36 times, the highest level since 2008; the percentage of buy ratings was below
60%; Among the Magnificent Seven, Apple is the only company with an implied return below the sell-side target price. A recent rally since the end of June has added approximately $1 billion to Apple's valuation.
A market capitalization of trillions of dollars reflects investors' bets on outstanding performance.
At this starting point, the margin for error is nearly zero. Outperforming the top line is the bare minimum, not a bonus; each of the three factors—Greater China earnings surprise, service growth deceleration, and weak gross margin guidance—has its marginal pricing weight amplified by valuation multiples. More notably, the direction is asymmetric: revenue guidance (+7%
At 10% (compared to consensus at 6.1%), the figure actually exceeded expectations, but failed to lift the stock price—indicating that the market is truly pricing in the quality of growth, not just its magnitude.
There’s also a more technical reason at play: the market may have been modeling Apple using the wrong calendar. Bank of America analyst Wamsi Mohan on October 30
The report explicitly stated that Wall Street made "modeling errors"—Apple is releasing the iPhone 18 Pro, Pro Max, and Duo in September, while the base model and iPhone Air are slated for next year.
In March, the model习惯ually groups all generations of a product into a single table. Based on this, Bank of America lowered its fiscal 2027 iPhone shipment forecast by 9 million units to 247 million, compared to the Wall Street consensus of approximately 260 million.
billion), with expected revenue of $113 billion and EPS of $1.97, both below consensus.
The results show that Bank of America’s caution for this quarter was overly conservative—Apple’s actual revenue exceeded its forecast by approximately $1.5 billion. However, Mohan’s assessment points not to this quarter, but to December and March.
Modeling revenue and gross margin for the monthly and quarterly periods. Part of the post-market volatility reflects the market repricing the "analyst model error" itself.
Tenu's Debut: Signal Gaps Caused by Shifts in Product Timing
This is the first earnings report since John Tennus took over as CEO on September 1, and his first time hosting the earnings call.
His background shapes how the market interprets things: mechanical engineering background, 25 years at Apple, led the transition of Mac to Apple Silicon.
Migration. Within weeks of taking office, he pushed to reduce dependence on the spring-fall release cycle, eliminated several middle-management roles, and shortened reporting hierarchies—one unintended consequence of this organizational restructuring being that the earnings rhythm itself became harder to model: for decades, revenue curves rose during the holiday season and dipped by late summer; now, Apple aims to flatten this pattern, at the cost of depriving the market of its seasonal extrapolation tool.
This quarter marks the first time the market has been forced to confront this situation. The earnings report simultaneously features three timing mismatches: the iPhone 18 Pro was included for only about nine days; the standard iPhone 18 and the new iPhone
Air is not until the first half of 2027; the foldable iPhone Duo goes on sale on October 23, so it is not included in this quarter's financial report.
Duo thus became a "silent evidence." Data from third-party bloggers and analytics agencies show that Duo recorded approximately 12.5 million in sales within just three days of its launch in W43 (October 19–25).
One million units activated, ranking fifth on the weekly sales chart for that week; during the same period, the iPhone 18 Pro Max sold approximately 724,000 units and the iPhone 18 Pro sold approximately 335,000 units.
Ten thousand units. According to Morgan Stanley, the online pre-order wait time for certain models is approximately 14 weeks—the longest in their tracking history over the past decade; J.P. Morgan estimates that, due to supply constraints, Duo shipments in 2026 will reach approximately 8 to 9 million units.
Ten thousand units.
But none of these figures appear in the report. The market can only infer them from management’s tone—this is precisely why the signaling weight of this earnings call exceeds that of confirmed numbers.
Tenu provided only qualitative information. He stated that a "very small percentage" of Duos were returned due to display issues, the company is replacing them, and it is currently impossible to determine when supply and demand will balance: "Duo
It is currently restricted, but ramp-up progress is going well; we are working hard to deliver to more customers as soon as possible, but we cannot yet predict when supply and demand will balance. He also confirmed that memory costs for the September quarter were higher than those for Q6.
The quarter was offset in part by lower inventory costs and reduced costs for some non-memory components, while the $100 price increase for the iPhone 17 and price hikes on select MacBooks "provided some hedging." He said the new model
Siri now handles 25 billion requests per day.
Thus, the pricing logic for this night was inverted: all confirmed figures (revenue, EPS, iPhone, Mac) exceeded expectations, but the stock price was determined by an unconfirmed figure—whether the gross margin for the December quarter could be maintained.
46%, when will Duo supply ramp up, and will the release timeline for the base iPhone 18 continue to disrupt the model?
Conclusion: Defense comes from "not spending," but "not spending" is not a growth engine.
Apple's current core tension is clear: it is being passively reclassified as a defensive asset due to its refusal to participate in the AI capital expenditure race, yet AI
The expansion of computing power is eroding its profit margin through increased storage costs; its defense stems from restraint, not from a new growth engine; and this type of defensive funding is tactical and reversible.
Growth gaps and cost pressures appear on the same report—Greater China and Services have created forecast variances for two consecutive quarters, and gross margin guidance for the first time includes both storage and AI costs in its forward-looking口径. At 36
At a P/E ratio double expectations and the highest valuation since 2008, "safety" needs to be revalidated, not assumed.
Tenes currently has a window of opportunity for a defensive narrative: the holiday season product lineup is the most密集 one in recent years (touch-bar MacBook Pro, updated iMac, new entry-level MacBook Pro).
Shipped on November 6 (HomeView home hub and AirPods 5 now available); Siri delivered; Duo
Demand signals are strong, but supply is constrained—this supply constraint itself can be framed narratively as "demand being too strong." How long this window remains open depends on several indicators: 12
For the current quarter in Greater China, will the expected discrepancy continue? Will service growth stabilize within the guidance range of 8% to 9%? Will the gross margin guidance confirm that storage pressure extends into fiscal year 2027? Apple’s AI
Has the capital expenditure口径 changed, and can Duo’s demand and supply guidance be shifted from “unpredictable” to “providing specific time points”?