Apple.
What the price already holds.

Purpose. This is a quality sample of company research. It states findings and their evidence.

Fair value. Our central estimate is $167.05 per share. The 21 Oct 2026 close of $343.99 is 105.9% above this value.^S12S13 Two practitioner checks give $137.37 to $267.40 (cross-check).

Risk measure. Scenarios use real company years. The FY2023 trough regime gives $126.06. The 9M FY2026 run-rate regime gives $211.74. The range is 51% of the central value. The price is 62.5% above the run-rate case.

The business. Apple is an exceptional franchise. Over the twelve months to June 2026 it earned 33.2% operating margin on $467bn of revenue, and Services carried a 76.0% gross margin.^S2S3 Capex is small: 2.2% of revenue. Free cash flow was $137bn.^S2S3 The question is not quality. It is how much growth the price already holds.

What changes the view. The discount rate moves value most: one point off the WACC gives $196.27. Services growth is second. Four thesis breakers sit outside the weights (thesis breakers): the loss of the Google search payments gives $153.72; a 30% fall in Greater China sales gives $153.76; a one-third cut in App Store commission gives $155.99; a longer iPhone replacement cycle gives $152.61. Watch Services growth, Greater China sales and the appeal in the search case.

Review status: provisional. One blind round ran with seven separate reviewers. It closed three resolvable bridge items (unrecognised tax benefits, finance leases, the August dividend) and corrected the cost of debt; the central value moved from $168.48 to $167.05, under 1%. A new resolvable fact returns the work to data collection, so a second quiet round is still needed. Open items: the nature of the $13.3bn intangible addition, the current size of search payments, and Q4 FY2026 results on 2 November.

Thesis intact. No results filing followed the Q3 FY2026 10-Q of 31 July. The CEO changed, the top court agreed to hear the App Store contempt question, and the discount rate stayed high. Each event sits in a model input or a thesis breaker; none moves a regime case.

Channels: estimate, multiple, discount rate, flows, read-across. An event without a mechanism is sentiment and does not enter the value.

Two practitioner checks sit beside the cash-flow value. Each uses Apple’s own history and reported figures.

Where they disagree. The P/E check is +60.1% from the DCF. Apple’s own P/E rose from 22.2× to 34.5× while the 10-year Treasury yield rose; the median therefore carries a multiple the discount rate no longer supports.^S10S12 The owner-earnings check gives growth no credit beyond 3%. At the price, the core P/E is 39.9×. All three methods put the value below the price.

Apple Inc. sells iPhone, Mac, iPad, wearables and services worldwide. It reports five geographic segments with the same products; the economics differ by category, not by region.^S3 The security is common stock (AAPL, Nasdaq), CIK 320193.^S8

Language models write fluently but compute and recall probabilistically. This method confines them to reasoning. Data and arithmetic stay outside the model.

Company class. A classification step runs before data collection. Apple is a profitable consumer company with a long history and no material goodwill, so it takes the profitable-company route: the Faustmann ratio with price to tangible book beside it, after-tax ROIC against the cost of capital, then owner earnings and a reverse DCF. The register gives SIC 3571, which maps to no sector; the 10-K business description sets the sector to consumer.^S3S8 The five geographic segments share one product set, so the model splits value by Products and Services, where margins differ.

Two-phase review. Phase A closes resolvable facts from primary data before any judgement round: the statutory tax rate, the dated 10-year yield, a regression beta and a dated equity premium. Phase B runs blind rounds. A result is stable when two rounds in a row keep the conclusion, move the central value by less than 1% and find no new resolvable fact.

Review status: provisional. One blind round ran with seven separate reviewers. It closed three resolvable bridge items (unrecognised tax benefits, finance leases, the August dividend) and corrected the cost of debt; the central value moved from $168.48 to $167.05, under 1%. A new resolvable fact returns the work to data collection, so a second quiet round is still needed. Open items: the nature of the $13.3bn intangible addition, the current size of search payments, and Q4 FY2026 results on 2 November.

Signal strength. Strong: audited and filed figures (levels 1–3). Medium: dated market data and practitioner datasets. Weak: press reports and trackers (section 11).

Weights. Scenario weights are the reader’s judgement. The calculator in the Summary shows how the weighted value moves. This makes the judgement visible.

Compared with typical AI research. Typical output states numbers from memory and one valuation. Here every figure carries a source and date, the arithmetic is reproducible, and the conclusion faced blind challenge.

At central margins, capex and WACC, the price is reached only when every growth rate in the central path is multiplied by 2.50. Section 08 shows the full solve and two matrices.

Implied revenue. The price requires about $1,388bn of revenue a year in fiscal 2035, in nominal USD. Our central path reaches $716bn.

Implied growth. That is 12.8% a year for ten years, against 5.6% in the central path and about 1.8% a year over FY2022–25.

Implied discount rate. Alternatively the price fits central operations at 6.08%, about 3.4 points below our dated 9.47% and below the 5.13% 10-year Treasury yield plus any equity premium above about 1 point.

Implied market share. On the IDC path the smartphone market reaches about $626bn in 2035. The implied iPhone revenue is 71% of it, against 38%–47% in 2025. If Services stay on the central path and Products carry the whole gap, iPhone revenue would need 2.2× the 2035 smartphone market, which is impossible. Services at a 100% gross margin with central growth give only $209.14.

Market-size estimates used

Market growth rates beyond the dated estimates are our assumptions. Effect on value: none; this tests the price, not the model.

Each scenario starts from the Base path and changes one mechanism. They are stress tests, not weighted cases. Loss sizes are our assumptions.

Reading. The price is $176.94 above Base. Each breaker widens that gap by $11.07 to $14.44. All four together give $115.98. The gap comes from growth and the discount rate, not from these risks.

Apple’s asset base is light, so its fallback is not renting assets. It is returning all free cash flow through buybacks and dividends.

Analogues.

• IBM, FY2011 to FY2018: revenue fell from $106.9bn to $79.6bn while diluted shares fell from 1.214bn to 0.916bn after $71.1bn of buybacks in FY2010–14. The share price fell from $175.79 (Dec 2011) to $108.67 (Dec 2018). Buybacks at a full multiple did not protect holders.^S40S41
• Apple itself, from the March 2012 capital return start: $816bn of repurchases in FY2013–25 cut split-adjusted diluted shares from 26.1bn to 15.0bn, while net income rose. Buybacks worked because they were bought at lower multiples.^S42S9

Conclusion. With no growth, the business is worth about $91.54 per share, and the price is 275.8% above it. At the price, all free cash flow retires about 2.5% of shares a year. Buybacks raise per-share figures, but at a price above value they move value from continuing holders to sellers.

The two markets behind Apple’s revenue grow slowly in units and faster in price. The price needs share gains on top of market growth.

Thesis. Apple shipped the most smartphones in 2025 and already takes about two-fifths of industry revenue.^S22S24 Its installed base passed 2.5bn active devices.^S36 Price rises such as the $100 step on iPhone 18 Pro lift revenue without new units.^S34 Services monetise the base, not new shipments.

Counter-arguments.

• Unit growth is near zero. IDC sees 1.6% a year to 2029 and a falling average price; revenue share can rise only by taking premium buyers from rivals who also raise prices.^S25
• Services already equal about two-thirds of the whole 2025 in-app market. Commission rules in the US, EU and Japan cap how much more of that spend Apple keeps.^S1S26
• Search licensing is a payment from one partner under an antitrust order. It grows with that partner’s search revenue, not with Apple’s devices.^S1S17

Conclusion. The markets can grow, but not fast enough. The price needs iPhone at about 71% of smartphone revenue in 2035, or Services far above the size of the app economy. No dated estimate we found supports either.

Apple holds no material listed or private investees. These are the counterparties and rivals named in this note.

Listing months are as commonly reported and were not checked against exchange records (Unverified); they set status, not value. No entity here enters the bridge.

AAPL is Apple Inc. common stock on Nasdaq. The SEC register lists SIC 3571.^S8 The fiscal year ends on the last Saturday of September.

The latest periodic filing is the Q3 FY2026 Form 10-Q, filed 31 July 2026. No FY2026 10-K is on file on 22 October 2026.^S8 The company set results for 2 November 2026.^S35 John Ternus became CEO on 1 September 2026.^S7

Products earn most of the revenue. Services earn twice the gross margin and grow faster, so their share of gross profit rose from 33% in FY2022 to 40% TTM.

Reported by the company; TTM = FY2025 + 9M FY2026 − 9M FY2025, calculated by us.^S2S3S4

Tariff refunds. Q3 FY2026 gross margin included about 2 points of tariff refunds, and EPS included $0.11.^S2 We estimate the pre-tax refund at $2.0bn and remove it from core earnings. TTM Products gross margin without it is 38.6%.

Opex is rising. R&D grew 33% in 9M FY2026.^S2 The model holds opex at 15.5% of revenue in every case, the 9M FY2026 level; FY2025 was 14.9%. SBC ($13.7bn TTM) sits inside opex and is treated as a cost.

Customers. No customer is named above 10% of net sales in the FY2023–25 10-Ks. Carriers held 34% of trade receivables at FY2025 year-end, and one customer held 12%.^S3S5 The 10-K does not name that customer; we searched FY2023, FY2024 and FY2025 10-Ks (Unverified).^S3S4S5

Competitor shares. In 2025 Apple shipped 19.7% of smartphones (IDC) or 20% (Counterpoint); Samsung 19.1% or 19%. Both are press reports of preliminary tracker data.^S22S23

Capex is about 2–3% of revenue and close to D&A. Suppliers fund the working capital.

Maintenance versus growth. The company does not split capex. On a D&A proxy, TTM maintenance capex was about $13.1bn and growth capex was −$3.1bn, so capex ran below the proxy. Depreciation of PP&E alone was $8.0bn in FY2025.^S3 The model uses capex 2.8% and D&A 2.6% of revenue in all cases.

Purchase obligations. Manufacturing purchase obligations were $57.0bn at 27 June 2026, $56.2bn due within 12 months; other purchase obligations $29.3bn; unconditional obligations $27.6bn.^S1 The filings searched (10-K FY2023–25, 10-Q Q2–Q3 FY2026) report no amount of write-downs on these obligations; we run no loss stress (Unverified).^S1S3S4S6

Compute. Apple relies on its own data centres and on third-party cloud providers for AI compute, and warns of constrained supply and rising cost.^S1 A shift to heavy AI capex would change this profile; nothing in the filings shows it yet.

Net cash is small against the market value: $4.26 per share.

Unrecognised tax benefits and lease balances are from the FY2025 10-K.^S3 Operating leases ($12.5bn) are not subtracted because their cost sits inside opex.

Balances are reported by the company.^S2 The Q4 FY2026 flow is our estimate (Products +10%, Services +13% on Q4 FY2025); it will be replaced by the 10-K. Unexplained intangible. Gross intangible assets rose $13.3bn in 9M FY2026 while other non-current liabilities rose $13.5bn; cash spent on acquisitions was small. The filings do not say what was acquired (Unverified).^S1S6 Treating the liability rise as debt gives $166.13.

The share count is the balance-sheet count, one document, one date. Later buybacks and Q4 cash use are inside the Q4 flow estimate. Unvested RSUs are excluded because their cost is inside opex.

All rate inputs are dated and closed from data before any judgement round.

Terminal growth is 2.5% / 3.0% / 3.5% by case, below the risk-free rate. A US premium only is a simplification for a global issuer.^S39

Each case copies a regime Apple has lived through. No case pushes every factor at once.

Bull Products growth starts below the 9M FY2026 rate because that rate includes a launch cycle and price rises; it keeps the regime’s margins. Opex stays at the 9M FY2026 level in every case, above the trough-year ratio, because R&D now grows faster than revenue. Returns on new capital sit far below the measured incremental return (section 09) to stay conservative. All forecast inputs are our assumptions.

Plausibility. Tangible book is $5.62 per share, so tangible-book floors do not bind for this business.

Central value $167.05 per share. Services carry 59% of it on 26% of TTM revenue.

Cash flows sit at mid-year from FY2027 and are discounted to 21 October 2026. The segment split charges opex pro rata to revenue; the total does not depend on it.

Central path, calculated by us. FY2026 revenue (9M reported plus Q4 estimate) is $478bn.

At central margins and WACC, the $343.99 price requires revenue to grow 12.8% a year from FY2025 to FY2035. The central path has 5.6%.

Revenue would rise from $416bn in FY2025 to about $1,388bn in FY2035. Alternatively, the price fits central operations at a discount rate of 6.08%.^S38 Services at a 100% gross margin on central growth give $209.14, so margin alone cannot close the gap.

Implied FY2025–35 revenue CAGR: WACC × Services gross margin

Central value per share: WACC × terminal growth

No cell of the value matrix reaches the price. The highest cell, WACC 8.47% and growth 4.0%, gives $220.62. Growth matters more than margin: a three-point Services margin step moves the implied CAGR by well under one point.

This is the screen for the class. Apple returns almost all its equity through buybacks, so book-based ratios are extreme.

NOPAT uses the 9M FY2026 effective tax rate for both years. The test passes on returns: every ROIC measure is far above the 9.47% WACC, and new capital since FY2023 earned about 98%. It fails on price: at 46.7× net worth and 61.2× tangible book, the market pays for future returns, not for assets. Invested capital is small because suppliers and customers fund working capital, so ratios on it overstate; the gross-asset ROIC of 87% is the steadier measure.

These items are context, not evidence for the value. They do not enter the model.

Every financial table states its period and unit. Reported figures keep issuer provenance. Each filing figure was read twice: from the filing text and from the XBRL register.^S9 Same-issuer filings and syndicated copies do not give independent confirmation. Forecast inputs are judgements.

This report contains public company research only. Figures may differ by rounding.

A. Scenario assumptions

B. Annual forecasts, $bn

C. Valuation outputs

D. Base inputs and provenance

E. Sensitivities on the central value