Johnson & Johnson JNJ

Reference table

1. What the business actually is

The 10-K describes two reportable segments. Innovative Medicine covers oncology, immunology, neuroscience, pulmonary hypertension, infectious diseases and cardiovascular/metabolism. MedTech covers surgery, orthopaedics, cardiovascular devices and vision. The consumer business is gone: J&J separated Kenvue and, per the 10-K, completed the debt-for-equity exchange of its retained stake in fiscal 2024 at a loss of $0.4 billion. There is no residual Kenvue financial effect in the current statements, and the word does not appear once in the narrative sections of the July 2026 10-Q. That question is closed.

Segment revenue, and it reconciles exactly.

The vendor seg_prod figures ($60,401M and $33,792M) sum precisely to reported revenue and match the 10-K's narrative ("Innovative Medicine segment sales in 2025 were $60.4 billion"; "MedTech segment sales in 2025 were $33.8 billion"). No override needed for FY2023 onward. For FY2022 and earlier the two sources do NOT agree and we say why in Section 8.

Innovative Medicine by therapeutic area, FY2025 — from the 10-K's MD&A, not the vendor file (the vendor carries no therapeutic-area detail):

MedTech by franchise, FY2025 — also from the 10-K:

Geography, FY2025 (vendor seg_geo, cross-checked to the 10-K): United States $53.752B (57.1%), Europe $21.535B (22.9%), Asia-Pacific and Africa $14.031B (14.9%), Western Hemisphere excluding the US $4.875B (5.2%). The 10-K states US sales of $53.8 billion and international of $40.4 billion; the vendor's non-US components sum to $40.441B. Consistent. The 10-K adds that three wholesalers represented approximately 21.8%, 15.5% and 11.1% of total gross revenues in 2025 — 48.4% of gross revenue through three customers, a concentration the vendor file does not disclose anywhere.

Headcount: 141,700 (profile). CEO Joaquin Duato. Listed on the NYSE since 1962.

2. The multiple everyone quotes is wrong, and the filings say by how much

Trailing four-quarter diluted EPS from the vendor's income statements sums to $8.63. At $254.93 that is 29.54x, which would make Johnson & Johnson one of the more expensive large-cap pharmaceutical companies in the world. It is not. The vendor's earn_cal block reports a completely different set of numbers for the same four quarters:

The filings settle which is which, and the 2026-07-29 8-K states the definition outright. In its footnotes to the updated guidance table, J&J defines adjusted EPS as a "Non-GAAP measure; excludes intangible amortization expense and special items." The 10-Q supplies the amortisation: intangible amortisation of $0.8 billion in Innovative Medicine and $0.5 billion in MedTech in the June quarter alone — $1.3 billion a quarter, roughly $5.2 billion a year, or about $2.13 per diluted share pre-tax. The 10-K puts fiscal 2025 intangible amortisation in cost of products sold at $4.6 billion. The "special items" add restructuring ($0.2B in the June quarter), Orthopaedics separation costs ($0.3B in the June quarter), acquisition and integration expense ($0.1B), and talc charges ($0.4B in the June quarter, $0.8B in the first half).

inc_q is GAAP. earn_cal is adjusted. The vendor labels neither. Analyst estimates are built on the adjusted basis — est FY2025 epsAvg is $10.803 against a trailing adjusted run-rate of $10.86, whereas GAAP FY2025 EPS was $11.03 and GAAP FY2024 EPS was $5.79. We therefore run every forward multiple in this dive on the adjusted basis and state both.

And GAAP annual EPS for this company is close to meaningless, which is worth spelling out because the vendor presents it without warning. Annual GAAP diluted EPS reads $13.72 (FY2023), $5.79 (FY2024), $11.03 (FY2025). Those swings are almost entirely talc and the Kenvue disposal:

• FY2023's $35.153B of net income in the vendor file is irreconcilable with the same row's pre-tax income of $15.062B less tax of $1.736B (= $13.326B, which is exactly the figure the cash-flow statement carries). The ~$21.8B gap is the Kenvue separation gain in discontinued operations.
• FY2024 carried, per the 10-K, "charges of approximately $5.1 billion for talc matters" plus a $0.4 billion loss on the Kenvue debt-for-equity exchange and $1.25 billion of acquired in-process research and development for the Yellow Jersey (NM26) rights. GAAP EPS of $5.79 is a distortion.
• FY2025 carried the mirror image: "the reversal of approximately $7.0 billion, a significant portion of the previously accrued talc reserve" in the first quarter, recorded at approximately a 22% tax rate. That single item is roughly $2.25 of the $11.03. It is why Q1 2025 shows net income of $10.999B and diluted EPS of $4.54 on revenue of $21.893B — a 50% net margin that never happened.

Verdict on the earnings basis: use the adjusted series, at 23.5x trailing and 19.8x FY2027E. Anyone screening this name on 29.5x trailing GAAP is looking at an artefact of litigation accounting.

3. The talc settlement, the patent cliff, and the three 8-Ks in eight days

3a. Three filings in eight days — what each one actually says

This is the analytically decisive section of the dive. We read all three.

2026-07-28 (event date 2026-07-27) — Item 7.01, Regulation FD. Verbatim:

> "the Company announced that it has reached an agreement for a comprehensive resolution of the ovarian talc litigation with the plaintiff firms leading the federal Multi-District Litigation (MDL) and related state court proceedings, conditioned upon, among other things, the express participation of at least of 95% of the remaining claims. The agreement calls for per claim payments, with a $5.5 billion commitment by the Company, and the first payment of no more than $3 billion to be made in 2027 and no additional payments due before 2028."

2026-07-29 — Item 7.01. Verbatim:

> "the Company announced the completion of the acquisition of Firefly Bio, Inc. ... and the entry into strategic agreements and collaboration with Sail Biomedicines ... The combination of these two transactions is expected to reduce the Company's adjusted operational earnings per share ... and adjusted earnings per share ... by approximately $0.64 in 2026, consisting of approximately $0.46 attributable to the Firefly acquisition and approximately $0.18 attributable to the Sail transaction."

and, further down:

> "The combination of these two transactions is also expected to reduce the Company's 2027 Adjusted Operational EPS and Adjusted EPS by approximately $1.36, consisting of approximately $0.08 attributable to the Firefly acquisition and approximately $1.28 attributable to the Sail transaction, which assumes Sail's achievement of certain development milestones and the Company's option grant, if exercised."

The 8-K explicitly states the Company "is updating its full year 2026 guidance relative to the guidance previously issued on July 15, 2026." Note the mechanics: the 10-Q discloses Firefly as a $1 billion cash transaction announced 2026-06-08. A $0.46 per-share 2026 hit on ~2.44 billion diluted shares is approximately $1.1 billion — i.e. essentially the entire purchase price is being expensed as acquired in-process research and development rather than capitalised. The $1.28 of 2027 Sail dilution implies roughly $3.1 billion of milestone and option payments. These are real cash commitments disclosed six days ago.

2026-08-04 — filed today, Item 5.02. Verbatim:

> "On August 4, 2026, the Company announced that Jennifer Taubert will retire from her position as Executive Vice President, Worldwide Chairman, Innovative Medicine and member of the Executive Committee effective September 1, 2026."

A same-day 8-K is a material event by definition, and this one is an Item 5.02 departure of the executive responsible for 64.1% of company revenue, with a four-week notice period and no successor named in the filing. We do not read a single retirement as a thesis-breaker — but we note it, we note that no successor is disclosed, and we note that it lands six days after the company committed $5.5 billion to a settlement and one week after it took $1.36 of 2027 earnings out of its own guidance. Three Item 7.01/5.02 filings in eight days is a cluster, and it is the reason this dive is written as a Hold rather than as a routine quality-compounder note.

3b. Talc — the number on the balance sheet versus the number in the press release

The 10-Q's contingency note (Note 11) is the primary source and it is unusually specific:

> "As of June 28, 2026, there are approximately 76,000 plaintiffs in the United States with direct claims against the Company and its affiliates in pending lawsuits regarding injuries allegedly due to the use of body powders containing talc, primarily JOHNSON'S Baby Powder."

> "As of the second quarter of 2026, the total present value of the reserve for talc related matters is approximately $3.7 billion, comprising executed settlement agreements, litigation defense and other costs. Approximately 40% of the reserve is recorded as a current liability."

The history is in the same note. Beginning October 2021 the Company undertook restructurings creating Red River Talc, LLC (ovarian and gynaecological claims) and Pecos River Talc LLC (mesothelioma, governmental and other claims), and filed three successive Chapter 11 proceedings. Per the filing: "Ultimately, each of the bankruptcy proceedings was dismissed and, as a result, the Company reversed substantially all, or approximately $7.0 billion, from amounts previously reserved for the bankruptcy resolution in the fiscal first quarter of 2025. Litigation in the tort system recommenced."

The reserve rollforward, from the filings:

The gap is $1.8 billion and it has to be recognised. The $3.7 billion reserve covers ovarian and mesothelioma and defence costs; the $5.5 billion commitment covers ovarian claims alone. We therefore expect a top-up charge at the Q3 2026 print on 2026-10-13, and we register that as a falsifiable prediction in Section 6c.

Two further talc threads remain open and neither is covered by the settlement. (1) A securities class action filed February 2018 in the District of New Jersey alleging failure to disclose asbestos contamination: class certified December 2023, affirmed by the Third Circuit July 2025, rehearing denied October 2025, certiorari denied by the Supreme Court in April 2026, with expert discovery now proceeding. (2) The Imerys/Cyprus global settlement: approved by the Delaware Bankruptcy Court in October 2024, insurers' appeal denied by the district court in August 2025, Third Circuit appeal fully briefed with a ruling pending.

And the settlement's own condition is live. In the MDL the court is addressing Daubert motions; in January 2026 the Special Master excluded certain plaintiff general-causation opinions while allowing others; in May 2026 evidentiary hearings were held on specific causation, after which plaintiffs withdrew two of their specific-causation experts; in June 2026 the Company filed a Motion to Show Cause asking why the MDL should not be dismissed; in July 2026 the court GRANTED the Motion to Show Cause, with a case-management conference scheduled for August 2026. That is the nearest real catalyst in this name, and it cuts both ways: a strong court outcome reduces the Company's need to pay $5.5 billion, which would be a positive surprise; a weak one jeopardises the 95% participation threshold.

3c. The patent cliff — Stelara now, Simponi and Opsumit next, Erleada in 2028

This is the house anchor thesis applied directly. The 10-K is blunt:

> "Sales of the Company's second largest product, STELARA (ustekinumab) accounted for approximately 6.5% of the Company's total revenues for fiscal 2025 ... The Company expects continued launches of biosimilar versions of STELARA globally which will continue to negatively impact the Company's sales of STELARA."

US Stelara sales were "approximately $3.8 billion in fiscal 2025." At 6.5% of $94.193B, worldwide Stelara was roughly $6.1 billion.

The erosion curve, measured as the drag on operational sales growth — this is the single most useful trend in the filings:

The drag is decaying, and it is decaying fastest internationally (−4.4% to −2.1%), which is where biosimilars launched first. That is the mathematically expected shape of a loss-of-exclusivity curve and it is the strongest single argument for the forward earnings path.

What is queued behind Stelara, all from the 10-K and 10-Q:

• Simponi / Simponi Aria: "At least two biosimilars are pursuing regulatory approval for a SIMPONI biosimilar in the United States, which would likely result in a significant reduction in future sales."
• Opsumit: "The Company expects generic competition for OPSUMIT in 2026, which would likely result in a significant reduction in future sales." Pulmonary hypertension was $4.4 billion in 2025.
• Zytiga: already lost exclusivity; named as an offset to oncology growth in 2025.
• Imbruvica: losing share to competitive pressure and hit by Medicare Part D redesign. Our knowledge base independently records that "BeiGene's Brukinsa is global BTK leader via superiority over J&J's Imbruvica in the ~$12B market."
• Erleada: ANDA litigation live on both sides of the border — a Canadian trial scheduled for August 2026 against Sandoz; a New Jersey suit filed June 2026 against MSN/Novadoz; a Hetero suit settled confidentially in May 2026. And CMS published the Selected Drug list for 2028 in January 2026, which includes Erleada.
• Carvykti: 2seventy bio filed in the Unified Patent Court (Brussels) in January 2026 alleging infringement of EU Patent No. 3 689 383, seeking damages and an injunction; the US government intervened as patent owner in May 2026; oral hearing March 2027. An injunction risk on the fastest-growing cell therapy in the portfolio is not trivial.
• Tremfya: Xoma filed in August 2025 alleging failure to obtain a licence in connection with Tremfya commercialisation; motion to dismiss denied December 2025.

Medicare / IRA exposure, stated precisely by the 10-K. The first CMS Selected Drug list included Xarelto, Stelara and Imbruvica, with government-set prices effective from 2026 for Part D drugs and 2028 for Part B. Critically: "CMS has indicated that, beginning in 2027, it will remove Xarelto and Stelara from the Selected Drug List, such that the products will no longer be subject to the IRA's minimum pricing provisions." So the IRA headwind on the two largest exposures is a 2026-only event that reverses in 2027 — a materially better outcome than the market generally assumes for this legislation, and it is a genuine positive we did not expect to find. Offsetting it, Erleada joins for 2028. Part D redesign is separately named as a drag on Erleada, Imbruvica, Invega, Opsumit and Uptravi.

What replaces the cliff. The 10-Q's own language on the first half of 2026: oncology "achieved operational sales growth of 16.1%" on Darzalex, Carvykti "share gains and site expansion", Tecvayli on the "U.S. TECVAYLI + DARZALEX FASPRO approval and expansion in the community setting", Talvey, Rybrevant/Lazcluze and Erleada. Neuroscience "achieved operational growth of 13.4%" on Spravato demand and Caplyta's adjunctive major-depressive-disorder launch. And within the declining immunology line, growth of Tremfya "due to share gains across all indications with significant IBD launch momentum" plus Imaavy (nipocalimab) and Icotyde (icotrokinra) in the US — the two named successors.

And the pipeline is being bought as well as grown. In the first half of 2026 J&J closed Halda Therapeutics for $3.05 billion net of cash, of which the 10-Q records $2.8 billion as acquired in-process research and development and $0.6 billion as goodwill. Our knowledge base flagged this transaction in November 2025 — before the close — as "attractive, fits its prostate franchise and offsets upcoming patent expiries." That is a rare case of the tracked lane and the filing corroborating one another on a specific, dated, checkable fact. Add Firefly Bio (degrader antibody conjugates, ~$1 billion, closed 2026-07-29) and Sail Biomedicines, and the 2025-2026 deal book runs to roughly $22 billion including Intra-Cellular at $14.5 billion.

4. Growth and margins — including a gross-margin series we do not believe

Annual, six years. Note the 2023 discontinuity: FY2022 and earlier are presented on a continuing-operations basis excluding the consumer business, which is why revenue falls from $94.9 billion of segment sales in FY2022 to $79.990 billion of reported revenue.

The FY2025 gross margin of 72.8% is an artefact and the 10-K contradicts it directly. The filing states, of fiscal 2025: "Cost of products sold increased as a percent to sales", driven by "unfavorable product mix driven by the decline of STELARA sales" and "tariffs, unfavorable transactional currency and macroeconomic factors in the MedTech business." The vendor shows cost of products sold FALLING from 30.9% of sales to 27.2% — a 370 basis point move in the opposite direction.

We reconstructed the cause. In FY2024, gross profit less R&D less SG&A equals operating income exactly ($61.350 − $17.232 − $21.969 = $22.149B). In FY2025 it does not: $68.558 − $14.665 − $23.676 = $30.217B against reported operating income of $25.596B, a gap of $4.621 billion — which is, to within rounding, the $4.6 billion of intangible amortisation the 10-K places inside cost of products sold. The vendor moved amortisation below the gross-profit line in FY2025 and did not do so in FY2024. On a like-for-like basis, fiscal 2025 gross margin is (68.558 − 4.621) / 94.193 = 67.9%, DOWN from 69.1% — which is exactly what the 10-K describes. The vendor reports X; the 10-K shows Y; we use Y.

The same defect infects the 2026 quarters, and more visibly:

The adjustment adds back the intangible amortisation the 10-Q discloses by segment — $0.8 billion in Innovative Medicine plus $0.5 billion in MedTech, $1.3 billion in the June quarter — to cost of products sold. On that basis Q2 2026 gross margin is approximately 68.0% against 67.9% a year earlier: flat. A flat gross margin in a quarter with Stelara mix pressure and MedTech tariffs is credible. A 520 basis point expansion is not. Operating margin is unaffected by the reclassification and is the series we use: 28.3% in Q2 2026 against 28.7% a year earlier.

Segment profitability is where the real story is, and the vendor carries none of it.

MedTech is 35.9% of revenue and roughly 17% of segment pre-tax profit. Using the 10-Q's own figures — first-half revenue $49.4 billion of which Innovative Medicine was $31.8 billion — Innovative Medicine contributed approximately $11.6 billion of segment pre-tax income and MedTech approximately $2.4 billion. The 210-basis-point fall in MedTech's margin is attributed by the filing to Orthopaedics separation costs of $0.4 billion, increased R&D, increased commercial investment and "tariffs included in Cost of products sold." That is the clearest tariff disclosure we found in this batch.

The Orthopaedics separation is the structural response. Announced October 2025, targeted for completion 18 to 24 months after announcement — i.e. April to October 2027 — with the 10-K conceding the Company "intends to explore multiple paths to effect the planned separation." The business to be separated is named DePuy Synthes, and the 10-K records a consequential accounting change: trademarks associated with the DePuy Synthes brand were reclassified from definite-lived to indefinite-lived, reflecting "management's revised expectations regarding the future economic life and continued use of these trademarks through and following the planned separation." A new risk-factor section, "Risks related to the planned separation of our Orthopaedics business", appears in this 10-K and did not exist a year ago — that is the changed risk factor this dive was asked to find.

Consensus forward path (analyst counts in brackets). Revenue $101.272B FY2026 (16), $108.583B FY2027 (17), $116.933B FY2028 (18), $124.853B FY2029 (9), $134.042B FY2030 (9). Adjusted EPS $11.684 (12), $12.875 (12), $14.340 (5), $16.138 (2), $17.954 (2). The FY2029 and FY2030 EPS rows rest on two analysts each and are used for nothing. The FY2028 row's dispersion is extreme — a low of $10.759 against a high of $16.579 on five analysts — and we treat it only as a cross-check, never as an anchor.

5. Balance sheet and capital returns

Net debt: the filing confirms our recompute and contradicts the vendor. The vendor reports netDebt of $28.224B for FY2025 — total debt of $47.933B less cash and cash equivalents of $19.709B, discarding the $393M of short-term investments. Counting them:

> Cash and short-term investments $20.102B − total debt $47.933B = net debt of $27.831B.

The 10-K states it independently: "In 2025, net debt (cash and current marketable securities, net of debt) was $27.8 billion compared to net debt of $12.1 billion in 2024. Total debt represented 37.0% of total capital ... Shareholders' equity per share at the end of 2025 was $33.86." The vendor reports $28.224B; the 10-K shows $27.8B; we use $27.8B. The error is only $393 million — 1.4% — but it is the same structural defect that produces catastrophic misstatements elsewhere in this feed, and it is worth knowing that it is present here too.

The debt footnote checks out. The 10-K: "Total borrowings at the end of 2025 and 2024 were $47.9 billion and $36.6 billion, respectively" — matching the vendor's totalDebt of $47.933B and $36.634B exactly. The increase is explained: "The Company issued senior unsecured notes for approximately $9.2 billion" in the first quarter of 2025, used "to fund the Intra-Cellular Therapies, Inc. acquisition for approximately $14.5 billion." The current portion of long-term debt was $2.0 billion, with the remainder of short-term debt being commercial paper — $6.5 billion at a 3.81% weighted average rate and approximately two months' weighted average maturity. Liquidity backstops: a $12.5 billion 364-day credit facility secured June 2026 (expiring 2027-06-23, up from $10 billion a year earlier) and capacity to issue up to $20 billion of commercial paper.

Lease double-count: checked and CLEAN. The vendor reports capitalLeaseObligationsCurrent and capitalLeaseObligationsNonCurrent as zero in all three years, which in this feed is usually a red flag. Here the 10-K vindicates it: "Commitments under finance leases are not significant." The operating lease right-of-use asset was $1.3 billion and the lease liability $1.4 billion in fiscal 2025, with operating lease costs of only $0.2 billion. There is no hidden lease leverage in this name — a striking contrast to the technology names in this batch — and no double-count to correct.

Enterprise value. km_ttm.enterpriseValueTTM reports $642.969B. Rebuilt (market cap $614.355B + total debt $47.933B − cash and short-term investments $20.102B) it is $642.186B — a +0.1% residual, the cleanest EV in this batch. Using the June-quarter balance sheet instead (the 10-Q's "approximately $20.8 billion" of cash and marketable securities against "approximately $49.0 billion" of notes payable and long-term debt, a stated net debt position of $28.2 billion), EV is $642.6B. All three roads lead to about $642 billion and we accept the vendor figure.

Balance-sheet composition, and one uncomfortable number.

Half the balance sheet is goodwill and intangibles, and tangible book value per share is NEGATIVE at −$4.81. Book value per share is $34.82 (the 10-K says $33.86 on a period-end share count; both are correct on their respective bases), so the stock trades at 7.32x book. We do not think book value is a useful lens for a pharmaceutical company whose principal assets are patents and pipelines, and the vendor's price-to-book sub-score of 1 should be read as a mechanical artefact of that, not as a warning. Return on equity is 25.7%, return on invested capital 13.9%, return on capital employed 18.0%, and return on tangible assets 20.2% — those are the numbers that matter.

Capital returns — the dividend is excellent and the buyback barely exists.

The 10-K: "The Company increased its dividend in 2025 for the 63rd consecutive year. Cash dividends paid were $5.14 per share in 2025 and $4.91 per share in 2024." The Board declared $1.30 per share in January 2026, then raised the quarterly rate to $1.34 per share on 2026-04-14, and declared $1.34 again on 2026-07-15.

The buyback is not a buyback. Both the 10-K and the 10-Q describe repurchases in identical language: "Common stock purchases on the open market are made as part of a systematic plan to meet the needs of the Company's compensation programs." In the fourth quarter of 2025 the Company repurchased 9,722,756 shares; in the second quarter of 2026, 970,250 shares — a 90% reduction. No repurchase authorisation amount is disclosed anywhere in the extracted text of either filing, which is itself the finding: J&J is not running a capital-return buyback, it is running an anti-dilution mechanism, and it is not currently keeping up. Diluted shares have RISEN from 2.419 billion in Q2 2025 to 2.440 billion in Q2 2026 — up 0.9%. Total shareholder yield is therefore approximately 2.1%, all of it dividend.

6. Cash flow

Free cash flow is remarkably stable at $17-20 billion and capital intensity is trivial — capex is 5.1% of revenue and 21.1% of operating cash flow. That is the structural advantage of a pharmaceutical business over the infrastructure-heavy names elsewhere in this batch, and it is why the dividend is safe at a 62.9% payout of free cash flow.

The interesting line is acquisitions, not capex. J&J spent $17.541 billion on acquisitions in FY2025 (principally Intra-Cellular at $14.5 billion) and $15.146 billion in FY2024 (principally Shockwave) — 89% of free cash flow in FY2025, funded with $9.637 billion of net new debt. In the first half of 2026 it added Halda at $3.05 billion; in July it added Firefly at roughly $1 billion plus the Sail commitments. This is a company that reinvests essentially all of its discretionary cash into external pipeline, and whose earnings quality therefore depends on how those assets are accounted for. The FY2025 free-cash-flow yield is 3.21% ($19.698B / $614.355B); the vendor's freeCashFlowYieldTTM of 3.05% is consistent. First-half 2026 operating cash flow was $11.1 billion per the 10-Q, with cash of $20.4 billion at 2026-06-28 against $19.7 billion at year end.

Income quality is sound: incomeQualityTTM of 1.13 (operating cash flow above net income), effective tax rate 16.5% trailing, and a first-half 2026 effective rate of 15.5% against 17.8% a year earlier — the fall attributable per the 10-Q to share-based-compensation benefits (2.9 points) and the absence of the prior-year talc reversal, which had been "recorded at approximately 22%." The 10-K separately discloses a one-time deferred-tax remeasurement cost of approximately $1.0 billion from US tax legislation (OBBBA).

7. Valuation — priced in or room?

At $254.93 (market cap $614.355B, EV ~$642.2B):

Vendor quality model. Rating B+, overall score 3, with sub-scores: discounted cash flow 4, return on equity 5, return on assets 5, debt to equity 2, price to earnings 2, price to book 1. The returns sub-scores are earned; the price-to-book 1 is meaningless for this asset class (Section 5); and the price-to-earnings 2 is almost certainly computed off the GAAP trailing figure of 29.5x rather than the adjusted 23.5x, which makes it a half-truth.

Peer context. The vendor peer set is the right one for once — Merck ($316.24B), AbbVie ($430.74B), AstraZeneca ($241.21B), Novartis ($291.32B), Amgen ($210.50B), Gilead ($167.95B), Pfizer ($144.82B), Sanofi ($102.48B), Novo Nordisk ($196.77B) and UnitedHealth ($370.26B, a payer rather than a manufacturer and not comparable). No peer multiples are supplied in the file, so no peer-multiple comparison is drawn. We note only that AbbVie, covered separately today, trades at a materially higher forward multiple on a materially worse balance sheet.

7a. What today's price assumes (the inversion)

At $254.93 — 19.8x FY2027 consensus adjusted EPS of $12.875 — the price embeds roughly these falsifiable claims:

• The talc liability is capped at approximately $5.5 billion and the 95% participation condition is satisfied. (Our reading of the 2026-07-28 8-K.) The stock has risen 13.7% in three months and the settlement was announced seven days ago; the reserve on the balance sheet is $3.7 billion. This is the most fragile assumption in the price. If participation falls short and the tort system resumes for a meaningful share of 76,000 plaintiffs, the discount that closed reopens.
• Adjusted EPS reaches ~$12.88 in FY2027, i.e. +19.2% over the trailing $10.86 in two years. (Consensus-derived, 12 analysts on each of FY2026 and FY2027.) We cannot verify that this consensus incorporates the $1.36 of 2027 dilution J&J disclosed on 2026-07-29 — the estimate file is dated 2026-08-04 but carries no revision timestamp. If it does not, the true FY2027 figure is nearer $11.52 and today's price is 22.1x, not 19.8x.
• Revenue compounds ~7.5% through FY2028 to $116.933B. (Consensus-derived: 16, 17 and 18 analysts.) That requires the Stelara drag to keep decaying (6.2% → 5.0% → 4.6% is the observed path), oncology to hold mid-teens operational growth, and the Simponi and Opsumit expiries to be smaller than Stelara's.
• The market keeps paying ~20x forward for a 7% grower. (Our reading.) J&J spent most of the last decade at a mid-to-high-teens forward multiple with the talc discount embedded. Today's 19.8x is the discount removed. There is no obvious mechanism for further expansion, and the street's own low target of $255 says a meaningful minority of covering analysts agree.
• MedTech's pre-tax margin troughs near 13.8% and recovers post-separation. (Our number, from the 10-Q segment note.) It fell 210 basis points year on year on tariffs, separation costs and R&D. The separation completes no earlier than April 2027.

7b. The return bridge (why the multiple moves)

Expected return over 12-24 months decomposes as: adjusted EPS growth (+8.2% FY2026, +10.2% FY2027 on consensus) + multiple drift (we assume a HOLD at ~20-21x, not expansion) + shareholder yield (2.1%, entirely dividend, since the diluted share count is rising).

Our base case explicitly does NOT assume multiple expansion, and that is the honest difference between this dive and the last twelve months of price action. The 52.4% twelve-month return was overwhelmingly a re-rating: revenue grew 6.0% and adjusted earnings roughly 8%, so approximately 40 points of the 52 came from the multiple. The catalyst for that re-rating — talc resolution — has now been announced. A re-rating that has been catalysed cannot be catalysed twice.

From $254.93, if adjusted EPS reaches the FY2027 consensus of $12.875 and the multiple simply holds at today's 19.8x, the price is $255 — flat, because 19.8x IS the FY2027 multiple today. Rolling forward one further year at an unchanged 19.8x on FY2028's $14.340 gives $284, or roughly +11% over two years plus about 4.2% of accumulated dividends. Our base of $270 requires 21.0x FY2027E — about one turn of expansion — and we justify that turn on the removal of a litigation discount rather than on any change in the growth rate. That is a modest, defensible assumption and it produces a modest, honest 5.9% of upside.

The bear case does not require multiple compression at all. It requires only that the FY2027 consensus has not yet absorbed the Sail dilution: $12.875 − $1.36 = $11.52, at 18.5x, is $213. The de-rating in the bear case is arithmetic, not sentiment.

7c. Variant perception (where we differ, what would surprise)

• We differ from the screening universe on the earnings basis, and in the direction that makes the stock look CHEAPER, not more expensive. Trailing GAAP EPS of $8.63 gives 29.5x. The adjusted basis — the one analyst estimates are built on, and the one the Company's own 8-K defines as "excludes intangible amortization expense and special items" — gives $10.86 and 23.5x. Every screen showing J&J at 29x is showing an artefact of $5.2 billion of annual intangible amortisation plus talc charges. Watchable number: the gap between inc_q and earn_cal EPS at each print; it has been $0.36-$0.68 a quarter.
• We differ from consensus on FY2027, and this is our sharpest disagreement. The Company disclosed on 2026-07-29 that Firefly and Sail "reduce the Company's 2027 Adjusted Operational EPS and Adjusted EPS by approximately $1.36." The vendor's FY2027 consensus is $12.875 on 12 analysts. $1.36 is 10.6% of that number, and the disclosure is six days old. If the consensus is stale, the honest forward multiple is 22.1x rather than 19.8x. This is the single most checkable claim in this dive. Watchable event: the FY2027 consensus figure over the next four to six weeks; a revision toward $11.5-11.8 confirms us, a stable $12.8-12.9 means the street has judged the Sail milestones unlikely to be achieved or the option unlikely to be exercised.
• We differ on the talc reserve and expect a charge. The 10-Q carries $3.7 billion of present-value reserve; the 8-K commits $5.5 billion for ovarian claims alone, before mesothelioma, defence costs and the still-live securities class action. We expect a top-up charge at the 2026-10-13 print. That would be a GAAP event, not an adjusted-EPS event — J&J classifies talc as a special item — but it is a real cash commitment beginning in 2027. Watchable number: the talc reserve line in the Q3 10-Q.
• We do NOT have a variant perception on the operating business, and we say so plainly. Our estimate of Innovative Medicine's trajectory is the same as the street's, our reading of the Stelara decay curve is the same as the filings' own, and our base fair value of $270 sits 3.8% BELOW the street consensus of $280.77. Our edge in this name is entirely in the earnings basis, the FY2027 disclosure gap, and the discipline not to chase a completed re-rating. That is a reason to hold, not to buy, and a reason to keep the position modest.
• Positive surprise that would force a re-rate upward: the MDL being dismissed or substantially narrowed following the August 2026 case-management conference on the Motion to Show Cause, which would mean J&J does not have to pay the full $5.5 billion. Alternatively, immunology returning to operational growth — the drag going from −4.6% to −2% or better — or MedTech's pre-tax margin recovering above 15% before the Orthopaedics separation. Watchable events: the August MDL conference; the immunology line in the Q3 10-Q; MedTech segment pre-tax margin.
• Negative surprise that would break the thesis: the 95% participation threshold failing, returning tens of thousands of claims to the tort system. Or the FY2027 consensus dropping by the full $1.36 without a corresponding fall in the multiple's denominator being anticipated. Or a Carvykti injunction from the Unified Patent Court following the March 2027 hearing. Or the Innovative Medicine leadership transition producing further senior departures. Watchable events: talc participation disclosure; FY2027 consensus revisions; the UPC docket; further Item 5.02 filings.

Synthos fair values

All three anchors are multiples of FY2027 consensus adjusted EPS of $12.875 (12 analysts), cross-checked against FY2026E of $11.684 (12 analysts) and, cautiously, FY2028E of $14.340 (5 analysts).

• Bear ~$213 — 18.5x a Sail-adjusted FY2027E of $11.52 ($12.875 less the $1.36 the Company itself disclosed). Cross-check: 18.2x FY2026E consensus; 19.6x trailing adjusted. The scenario: the 95% talc participation threshold is not met, the FY2027 consensus falls by the disclosed dilution, and the multiple returns to the mid-to-high teens the stock carried before the settlement. Note that $213 is below the 200-day average of $227.07 and above the 52-week closing low of $170.59. −16.4%.
• Base ~$270 — 21.0x FY2027E consensus of $12.875. Cross-check: 23.1x FY2026E; 18.8x FY2028E; 24.9x trailing adjusted. Sits 3.8% below the street consensus of $280.77 and 4.3% below the median of $282. The scenario: the settlement holds, the Stelara drag continues to decay, oncology compounds in the mid-teens, and the market pays one additional turn for a company without a tail risk. +5.9%, plus 2.1% of dividend.
• Bull ~$308 — 23.9x FY2027E. Cross-check: 21.5x FY2028E; 26.4x FY2026E. Marginally above the street high of $305. The scenario: the MDL is dismissed or narrowed and the $5.5 billion commitment is never fully drawn, immunology returns to growth in 2027, the Orthopaedics separation is executed as a clean spin at a premium multiple, and Carvykti plus the bispecific portfolio compound faster than modelled. +20.8%.

Base is 5.9% above spot; asymmetry roughly 1.3:1 to the upside (16% down, 21% up). That is not enough compensation to buy after a 52% run. Hold.

8. Knowledge base — 15 claims, thin, clean, and five months out of date

Tracked knowledge-base claims naming Johnson & Johnson: 15 raw hits, of which 15 are entity-level. Searched across all 51,928 records using six patterns: JNJ (word-boundary), Johnson & Johnson, Johnson and Johnson, Stelara, Darzalex and Tremfya. Matches came from Johnson & Johnson (13) and JNJ (2); the three product-name patterns returned zero hits, which is itself informative — the lane discusses this company by name, not by drug.

Nothing was discarded. There are no thesis-only or category-level mentions, no false positives, and no quarantined-attribution claims in this file. For a name of this size that is unusually clean, and we report it as a strength of the sample even though the sample is small.

Breadth is the problem, and we do not soften it. Five distinct sources: biotech_hangout (11 claims), andy_jassy_mgmt (1), no_priors (1), business_breakdowns (1), eli_lilly_ceo (1). Eleven of fifteen claims — 73% — come from a single source. The effective independent breadth is closer to three. Stance mix is 12 bullish / 3 neutral / 0 bearish, which reads as strong support but is better read as an absence of a bear lane: nobody in the tracked universe has argued the negative case for this company in the last three years.

Date range 2020-12-08 to 2026-02-27. The newest claim is more than five months old. The knowledge base therefore contains nothing about: the talc settlement (2026-07-28), the Firefly and Sail transactions (2026-07-29), today's Innovative Medicine leadership change (2026-08-04), the Orthopaedics separation announcement (October 2025 — outside the claim set entirely), or the 52.4% twelve-month re-rating. The lane is a pipeline resource, not a market resource, and we treat it as one.

One claim is management-lane and is half-weighted (skill 0.5):

> 2026-02-27 · management lane (skill 0.5) · bullish · conviction 65

> "A rotation is underway: investors trimming Mag7/top-5 exposure toward traditional names like Walmart, J&J, Exxon and Verizon trading near 52-week highs; the equal-weighted index was up ~6% while cap-weighted sat flat."

This is the only claim in the file that touches the price action, it is half-weighted, and it was right. In late February it observed J&J trading near 52-week highs on a rotation into defensives; the stock has continued higher since. We record it and we do not lean on it.

Where the lane is genuinely valuable is the therapeutic detail, and it corroborates the filings on specific dated facts:

> 2025-11-21 · biotech_hangout · bullish · conviction 72

> "J&J's ~$3B buy of phase-one Halda (HLD-0915 oral 'hold-and-kill' prostate drug) is attractive, fits its prostate franchise and offsets upcoming patent expiries."

The 10-Q filed 2026-07-23 records the Halda acquisition at $3.05 billion net of cash, with $2.8 billion allocated to acquired in-process research and development and $0.6 billion to goodwill. The claim named the price, the mechanism, the franchise fit and the strategic rationale — patent-expiry offset — months before the filing confirmed it. That is the single best piece of corroboration in this file.

> 2025-12-15 · biotech_hangout · bullish · conviction 70

> "J&J's Majestic-3 (teclistamab+Darzalex) was standout ASH data: 83% vs 65% 36-month OS versus triplet, rivaling CAR-T; FDA approval expected 2026 with manageable infection risk via IVIG."

The 10-Q independently attributes first-half oncology growth partly to "TECVAYLI driven by launch uptake and share gains from the U.S. TECVAYLI + DARZALEX FASPRO approval and expansion in the community setting." The approval the claim anticipated has happened and is contributing to revenue.

> 2025-12-15 · biotech_hangout · bullish · conviction 70

> "J&J/Legend's Carvykti is the biggest/fastest-growing cancer CAR-T with tremendous efficacy, but faces capacity constraints and toxicity (parkinsonism, neurotox, infection deaths)."

> 2025-11-13 · no_priors · bullish · conviction 70

> "Carvykti (J&J via Legend Biotech partnership) is a commercially successful multiple-myeloma cell immunotherapy — proof Chinese-originated data replicated and scaled in the US."

The 10-Q's language — Carvykti growth "driven by continued share gains and site expansion", and in the 10-K "share gains and capacity expansion" — confirms the capacity constraint the lane identified. The lane is also honest about the competition:

> 2025-12-15 · biotech_hangout · bullish · conviction 65

> "Gilead/Arcellx's anito-cel has a cleaner safety profile than Carvykti; filing fourth-line in 2026, it becomes preferred where labeled and gains share as it expands earlier."

> 2025-12-15 · biotech_hangout · bullish conv 70

> "BeiGene's Brukinsa is global BTK leader via superiority over J&J's Imbruvica in the ~$12B market; clinicians still start with covalent inhibitors over Lilly's non-covalent pirtobrutinib."

The Imbruvica claim is corroborated exactly by the 10-K, which attributes an oncology growth offset to "IMBRUVICA (ibrutinib) due to competitive pressures." That is a tracked source and a regulatory filing describing the same commercial loss.

On neuroscience, three claims that together explain a franchise the vendor file cannot see:

> 2025-08-27 · biotech_hangout · neutral · conviction 60

> "Spravato (esketamine) annualizing $1.2B despite two-hour visits and working in only 2 of 4 trials, framing a real commercial paradigm for psychedelics."

> 2025-08-21 · biotech_hangout · bullish · conviction 70

> "Spravato validates hard-to-administer psychiatric drugs — despite modest efficacy and burdensome dosing it sells >$1B and growing, potentially 3B+."

> 2025-08-27 · biotech_hangout · bullish · conviction 65

> "CNS is a hot M&A theme; J&J's Intra-Cellular buy is largest since Karuna/BMS, the fourth multi-billion CNS acquisition in a year."

> 2025-08-28 · biotech_hangout · neutral · conviction 55

> "The ~$10B+ J&J/Intra-Cellular deal was an outlier; $10B+ deals stay rare, but multiple $1-5B+ transactions are in the works"

The 10-K confirms the neuroscience segment at $7.8 billion, +10.1%, with growth attributed to Spravato demand and Caplyta from the Intra-Cellular acquisition — closed 2025-04-02 for approximately $14.5 billion. The lane called both the strategic logic and the subsequent deal cadence, and the last five months of J&J filings (Halda $3.05B, Firefly ~$1B, Sail) are precisely the "$1-5B+ transactions" the August 2025 claim predicted.

The remaining two claims are old and of limited use: a 2023-05-06 observation about consumer-goods habit formation which is now irrelevant post-Kenvue, and a 2020-12-08 claim about the single-dose COVID-19 vaccine, a product the 10-K now reports as a decline within infectious diseases.

Why Moderate rather than higher. The lane is clean, specific, technically literate and repeatedly corroborated by filings on dated facts — that is worth something real. But it is 15 claims from an effective three sources, has no bear case at all, and has been silent for five months through the most consequential period in this name's recent history. We use it for the pipeline and we do not use it for the price.

9. Data integrity — what we rejected from the vendor file and why

Nine items. Three are direct filing-versus-vendor contradictions, and in every one the filing wins.

1. FILING OVERRIDES VENDOR — the fiscal 2025 gross margin of 72.8% is an amortisation-reclassification artefact. The vendor reports gross profit of $68.558B on revenue of $94.193B (72.8%) for FY2025 against 69.1% for FY2024. The 10-K states the opposite direction outright: "Cost of products sold increased as a percent to sales", driven by unfavourable Stelara mix, transactional currency and MedTech tariffs. The reconciliation: in FY2024 gross profit less R&D less SG&A equals operating income exactly; in FY2025 it exceeds operating income by $4.621 billion, which is the $4.6 billion of intangible amortisation the 10-K places inside cost of products sold. On a like-for-like basis FY2025 gross margin is 67.9%, down from 69.1%. We use 67.9% and we reject the vendor's 72.8%. The same defect makes the Q1 2026 (71.5%) and Q2 2026 (73.1%) quarterly gross margins unusable; adding back the 10-Q's disclosed segment amortisation of $0.8B (Innovative Medicine) plus $0.5B (MedTech) puts Q2 2026 at approximately 68.0% against 67.9% a year earlier — flat, not up 520 basis points.

2. FILING OVERRIDES VENDOR — netDebt ignores short-term investments; the 10-K states the correct figure. Vendor netDebt of $28.224B is total debt ($47.933B) less cash and cash equivalents only ($19.709B), discarding $393M of short-term investments. Correctly computed it is $27.831B, and the 10-K says: "In 2025, net debt ... was $27.8 billion." The vendor reports $28.224B; the 10-K shows $27.8B; we use $27.8B. The error is small (1.4%) but structurally identical to failures that are catastrophic in other files, and enterpriseValueTTM carries it: rebuilt EV is $642.186B against the vendor's $642.969B, a +0.1% residual — which we accept.

3. FILING OVERRIDES VENDOR — the vendor payload contains no record of the $5.5 billion talc commitment, the $1.36 of disclosed 2027 EPS dilution, or the Innovative Medicine leadership change. All three arrived by 8-K in the eight days before this dive (2026-07-28, 2026-07-29, 2026-08-04) and none appears in any vendor field. The 2026-07-29 disclosure in particular may invalidate the FY2027 epsAvg of $12.875 on which every forward multiple in this dive rests, and we flag it as unresolved rather than pretending otherwise.

4. The est block's EBITDA, EBIT and net-income rows are internally impossible — REJECTED; epsAvg used exclusively. In every forward year, estimated net income EXCEEDS estimated EBIT: FY2026 $28.477B versus $24.910B; FY2027 $31.071B versus $26.708B; FY2028 $34.317B versus $28.762B; FY2029 $39.206B versus $30.710B; FY2030 $43.618B versus $32.971B. A company that pays interest and tax cannot earn more net income than operating income. By FY2030 the file implies net income of $43.618B on EBITDA of $44.530B — a 98% conversion. All EBITDA, EBIT and net-income estimate rows are rejected. Only epsAvg is used, and epsAvg cross-checks acceptably: FY2025's $10.803 against a trailing adjusted run-rate of $10.86.

5. Trailing GAAP EPS is not the multiple, and the vendor never says which block is which. inc_q carries GAAP diluted EPS summing to $8.63; earn_cal carries adjusted EPS for the same four quarters summing to $10.86. The 2026-07-29 8-K supplies the definition — adjusted EPS "excludes intangible amortization expense and special items" — and the 10-Q supplies the amortisation, $1.3 billion in the June quarter alone. We use the adjusted basis (23.5x) for valuation and report the GAAP basis (29.5x) as an artefact. See Section 2.

6. Annual GAAP net income and EPS are distorted by talc and by the Kenvue disposal to the point of being unusable. FY2023 netIncome of $35.153B cannot be derived from the same row's pre-tax income of $15.062B less tax of $1.736B ($13.326B, which is the figure cf_a carries) — the ~$21.8B difference is the Kenvue separation gain in discontinued operations. FY2024 carried ~$5.1B of talc charges, a $0.4B Kenvue exchange loss and $1.25B of acquired IPR&D. FY2025 carried a $7.0B talc reserve REVERSAL in the first quarter, which is why Q1 2025 shows a 50.2% net margin and $4.54 of quarterly EPS. We use operating income and adjusted EPS for all trend work and treat annual GAAP EPS as non-comparable. The knock-on: annual ebitda is also unusable — FY2025's $41.055B is pre-tax income including the $7.0B reversal plus D&A plus interest, and FY2024's $24.781B is depressed by the $5.1B charge. Trailing EBITDA (~$31.9B implied by evToEBITDATTM of 20.14x) spans a cleaner window and is accepted with the qualification stated.

7. seg_prod is not comparable to inc_a before FY2023. For FY2022 the segment table reports Innovative Medicine $52.563B + MedTech $27.427B + Consumer $14.953B = $94.943B, while inc_a reports FY2022 revenue of $79.990B — the difference being the consumer business, which inc_a presents as discontinued operations and seg_prod does not. From FY2023 onward the two agree to the dollar ($54.759B + $30.400B = $85.159B; $56.964B + $31.857B = $88.821B; $60.401B + $33.792B = $94.193B). We use FY2023-FY2025 only for segment trend work and say so wherever the earlier years appear.

8. Lease double-count: tested and CLEAN, confirmed by the filing. The vendor reports zero capital-lease obligations in all three years. The 10-K corroborates: "Commitments under finance leases are not significant", with an operating lease liability of $1.4 billion and a right-of-use asset of $1.3 billion. No adjustment. This is the only major name in this batch where the lease footnote produced nothing.

9. Two minor internal inconsistencies, flagged once and resolved. (a) quote.yearHigh of $274.90 and yearLow of $169.65 are intraday extremes; tech.hi52 of $267.24 and tech.lo52 of $170.59 are closing extremes. We use the closing series throughout, so the stock is 4.6% below its high, not 7.3%. (b) quote.sma50 ($245.20) and quote.sma200 ($226.43) differ slightly from tech.sma50 ($246.17) and tech.sma200 ($227.07); we use the tech series, which is computed from six years of closes. quote.price and tech.last agree exactly at $254.93.

Not defects, correctly reported: priceToBookRatioTTM of 7.32x and tangibleBookValuePerShareTTM of −$4.81 are arithmetically correct and reflect that goodwill and intangibles are 49.8% of total assets after Shockwave, Intra-Cellular and Halda. They are correct and they are not meaningful for a pharmaceutical company, and the vendor's price-to-book sub-score of 1 should be discounted accordingly. dividendPayoutRatioTTM of 60.0% is computed on the GAAP basis; on the adjusted basis it is 48.3%.

Non-equity tripwire: checked and clear, with one note. JNJ is common stock on the NYSE, average volume 8.5 million shares, dividend rising rather than fixed (63 consecutive annual increases), and a 52-week closing range of $170.59-$267.24 — a 57% band, far too wide for a preferred or a baby bond. The one item worth stating is the beta of 0.235, the lowest in this batch and low in absolute terms — that is a genuine characteristic of a defensive mega-cap healthcare name in a year when the index was driven by technology, not a data error, and it is corroborated by the six-month return of +10.5% against SPY's +11.1%.

10. Technicals — a 28-point win almost nobody framed as one

• Price $254.93, up 0.20% (+$0.52) on the day, on volume of 6.7 million against an 8.5 million average — a quiet session, notably quiet given a same-day 8-K.
• −4.6% from the 52-week closing high of $267.24; +49.4% above the 52-week closing low of $170.59. Maximum drawdown from peak over the trailing year is −4.6% on closing prices — the shallowest drawdown in this batch. (On intraday prices the range is $169.65-$274.90; we use closes.)
• Above both moving averages, moderately: +3.6% over the 50-DMA of $246.17 and +12.3% over the 200-DMA of $227.07. The 50-DMA sits above the 200-DMA by 8.4%, a well-established uptrend.
• RSI 59.4 — mid-range, not overbought. MACD +2.86. For a stock 4.6% off its high after a 52% year, a 59.4 RSI is a healthy reading: the advance has been gradual rather than parabolic.
• Relative performance, and the split that matters:

• Read that table carefully, because it is not the picture the headline number implies. The twelve-month outperformance of 28 points is real and is the second-best twelve-month relative in this batch. But J&J underperformed SPY over six months, and underperformed QQQ by 5.2 points. The gain is concentrated in months seven through twelve — roughly August 2025 to February 2026 — and again in the last three months on the talc settlement. There was a flat stretch in between. That is consistent with the re-rating being event-driven rather than momentum-driven, and it means the twelve-month figure will decay mechanically as the older period rolls off.
• Practical read. There is no entry urgency in either direction. The stock is 4.6% from its high with a 59.4 RSI, 3.6% above its 50-day average, and offers a 2.10% forward yield. The 50-DMA at $246.17 (−3.4%) is the first meaningful support and the 200-DMA at $227.07 (−10.9%) the second. Our bear case of $213 sits below both, which is the honest statement of how far this could fall if the talc settlement fails. Nothing about the chart says sell; nothing about the chart or the price says buy more.

11. Insiders — routine cashless exercises, one worth a sentence

(The file also contains the paired option-cancellation legs of each exercise — 11,925 and 11,129 options at $131.94 and $129.51 for Broadhurst, 10,000 at $115.67 for Wengel — which are the mechanical other side of the same transactions and are not separate decisions.)

Both are textbook cashless exercises and neither is a signal on its own. Wengel exercised 10,000 options struck at $115.67 and sold exactly 10,000 shares at $241.15, retaining 114,288 — she monetised the spread and kept a large position. Broadhurst exercised 23,054 options across two strikes and sold exactly 23,054 shares at $251.27, ending with 23,003 shares — she also monetised the spread, though her retained holding is far smaller.

One observation, stated carefully and without inference of impropriety. Broadhurst's sale executed on 2026-07-20, eight days before the 2026-07-28 talc settlement announcement, at $251.27 — approximately 1.5% below today's close. Option exercises at this level are ordinarily pre-scheduled under trading plans, the sale price was below the subsequent market, and the transaction pattern (exercise-and-sell-the-same-quantity) is the least discretionary form of insider sale there is. We record the timing for completeness and read no signal into it.

The absence in this table is worth more than the presence: through a twelve-month period in which the stock rose 52.4%, not one officer or director in the eight most recent filings made a discretionary open-market purchase, and not one made a discretionary open-market sale beyond the mechanical monetisation of expiring option value. That is neutral, and neutral is the correct reading.

12. Verdict, kill-criteria and flip conditions

Hold.

A very good business, a genuinely improved risk profile, and a price that has already paid for both.

What has actually changed, and it is important. On 2026-07-28 Johnson & Johnson announced a comprehensive resolution of the ovarian talc litigation with the leadership of the federal MDL and the related state proceedings — a $5.5 billion commitment, a first payment of no more than $3 billion in 2027, no further payments before 2028, conditioned on at least 95% participation. For a company that had three Chapter 11 vehicles dismissed, reversed $7.0 billion of reserve in a single quarter, and still faces approximately 76,000 US plaintiffs, that is the removal of the item that has capped this multiple for a decade. It is the explanation for a 52.4% twelve-month return against a 24.3% index, and for a stock that trades at 19.8x forward when it spent most of the last decade in the mid-teens.

What you are accepting, stated without softening. The $5.5 billion commitment exceeds the $3.7 billion present-value reserve on the June balance sheet by $1.8 billion, and we expect a top-up charge. The 95% participation condition is not yet met. The securities class action survived certiorari denial in April 2026 and is in expert discovery. The insurers' Imerys appeal is pending at the Third Circuit. And on 2026-07-29 the Company itself disclosed that Firefly and Sail reduce 2027 adjusted EPS by approximately $1.36 — 10.6% of the FY2027 consensus of $12.875 on which every forward multiple in this dive is computed, six days old, and we cannot confirm it has been absorbed. The Innovative Medicine segment head — 64.1% of revenue — retires on 1 September per an 8-K filed today, with no successor named. MedTech's pre-tax margin fell from 15.9% to 13.8% on tariffs and separation costs. Diluted shares are RISING. And the street, which has watched a 28-point outperformance, is 16 hold and 3 sell with a low target of $255 — the price it closed at.

Why Hold rather than Buy — Tactical. Base fair value of $270 is 5.9% above spot and 3.8% below the street consensus of $280.77. We have no variant perception on the operating business — our read of the Stelara decay curve is the filing's own, and our revenue path is the street's. Our edge is confined to the earnings basis (23.5x adjusted, not 29.5x GAAP), the FY2027 disclosure gap, and the discipline not to chase a re-rating that has already been catalysed. Six per cent of upside against a 16% bear case, on a name that has run 52%, with the settlement's participation condition still open, is a Hold. Own it if you own it. Take the 2.10% yield.

Why not Avoid. Because the business is genuinely good and the balance sheet is genuinely clean: $27.8 billion of net debt against $24.5 billion of annual operating cash flow, 0.90x net debt to EBITDA, 33.1x interest coverage, 25.7% return on equity, 5.1% capex intensity, a 0.235 beta, 63 consecutive years of dividend increases, and — uniquely in this batch — a lease footnote with nothing hidden in it. At $227 (the 200-day average) this is a straightforward buy. At $255 it is not.

Pre-registered KILL criteria — what would break the thesis:

• The 95% talc participation threshold failing, or the settlement being abandoned. This is the whole re-rating. A collapse returns the stock to a pre-2026 multiple.
• A talc charge materially above the $1.8 billion gap between the $3.7B reserve and the $5.5B commitment — i.e. evidence that $5.5 billion is a floor rather than a cap.
• The Stelara drag ceasing to decay. The path is −6.2% (FY2025) → −5.0% (H1'26) → −4.6% (Q2'26). A reading that stops improving, or worsens as Simponi biosimilars arrive, breaks the revenue bridge.
• MedTech pre-tax margin falling below ~12%. It has gone 15.9% → 13.8% in a year.
• A Carvykti injunction from the Unified Patent Court following the March 2027 hearing, or an adverse Erleada outcome in the Canadian trial beginning August 2026.
• Further senior departures from Innovative Medicine beyond today's Item 5.02, or a successor appointment that signals strategic change rather than continuity.
• FY2027 consensus adjusted EPS falling below ~$11.50 without a corresponding fall in the share price — that is the bear case arriving.

Pre-registered FLIP TO BUY criteria (upsize):

• A retracement to the $225-235 area (the 200-DMA is $227.07) with the talc settlement intact — this is the single most likely path to a full-size position, and it implies roughly 16-19% of base-case upside rather than 6%.
• The MDL being dismissed or substantially narrowed at or after the August 2026 case-management conference on the Motion to Show Cause, which would mean the $5.5 billion commitment is unlikely to be drawn in full.
• Immunology returning to operational growth — the drag improving from −4.6% toward −2% or better, confirming the Stelara trough.
• Confirmation that FY2027 consensus already reflects the $1.36 of Sail and Firefly dilution, which would mean the forward multiple is genuinely 19.8x rather than 22.1x.
• The Orthopaedics separation announced as a specific structure and date at a valuation implying MedTech is worth more apart than together.

Where JNJ fits in the Synthos Framework Portfolio. The defensive healthcare / diseases-of-aging sleeve, at a 1.5-2% target weight if entered on weakness toward the 200-day average, or a maintained existing position at current levels — not a new full-size purchase here. The house standing rule that mega-cap incumbents are guilty until proven innocent on a forward basis applies with force: J&J's twelve-month return is trailing performance driven by the removal of a discount, not forward acceleration, and the forward revenue growth rate is 7.2%. What earns it a place at all is that its franchise concentration — oncology $25.4B, immunology $15.7B, neuroscience $7.8B — sits directly on the age-incidence curves the anchor thesis is built around, at 19.8x forward earnings and a 0.235 beta. That is a portfolio-stabiliser role, not a return-generator role, and it should be sized as one.

On overlap within the batch. JNJ and AbbVie (also covered 2026-08-04) are both large-cap pharmaceutical companies facing loss of exclusivity on their largest immunology asset, and they are not redundant but they are close. AbbVie carries roughly $69 billion of debt against a near-negligible equity base, a vendor quality rating of C+ with an overall score of 2, and trades at a materially higher forward multiple. J&J carries $27.8 billion of net debt, 25.7% return on equity, a B+ rating, and a clean lease footnote. If only one is held, on today's prices and balance sheets it should be this one — but neither is a compelling purchase at these levels, and both are rated Hold. Logged as a tracked Synthos call (Hold) as of 2026-08-04 at $254.93.

Single biggest risk: that the talc resolution does not hold. The settlement announced seven days ago is conditioned on at least 95% of remaining claims participating, that threshold has not been demonstrated, the $5.5 billion commitment exceeds the $3.7 billion reserve, approximately 76,000 US plaintiffs remain, and two separate appellate threads — the securities class action and the insurers' Imerys appeal — sit outside the settlement entirely. The entire 28-point twelve-month outperformance rests on this being over.

Most fragile assumption in the price: that FY2027 consensus adjusted EPS of $12.875 is current. The Company disclosed on 2026-07-29 that Firefly and Sail reduce 2027 adjusted EPS by approximately $1.36. If the consensus in this file predates that disclosure — and twelve analysts, six days, is a plausible lag — the honest forward multiple is 22.1x, not 19.8x, and the base fair value falls from $270 toward $242. This is the cheapest thing in this dive to check and the most consequential thing to be wrong about.

Provenance & disclosures

• Traceability: 15 entity-level knowledge-base claims name Johnson & Johnson (15 raw hits; zero discarded; breadth 5 distinct sources; net conviction positive). Stance mix 12 bullish / 3 neutral / 0 bearish, spanning 2020-12-08 to 2026-02-27. Eleven of the fifteen claims (73%) come from a single source, so effective independent breadth is closer to three. One management-lane claim (skill 0.5) is half-weighted and identified where cited; it is not used as a valuation input. Searches covered six patterns — JNJ (word-boundary), Johnson & Johnson, Johnson and Johnson, Stelara, Darzalex, Tremfya — across all 51,928 records; the three product patterns returned zero hits. The lane contains nothing dated after 2026-02-27 and therefore nothing about the talc settlement, the Firefly/Sail transactions, the Innovative Medicine leadership change, the Orthopaedics separation or the re-rating — stated as a limitation, not smoothed over. All quoted claims are verbatim from the stored claim text.
• Data as-of: fundamentals — fiscal Q2 2026 ended 2026-06-28, 10-Q filed 2026-07-23, plus the fiscal 2025 10-K filed 2026-02-11 and three 8-Ks dated 2026-07-28, 2026-07-29 and 2026-08-04 (today) · estimates 2026-08-04 · prices 2026-08-04 ($254.93; 50-DMA $246.17; 200-DMA $227.07; RSI 59.4; MACD +2.86) · knowledge-base claims 2026-08-04. Every figure comes from the Synthos vendor data file for JNJ or from the SEC filings in filings/JNJ/; no figure comes from memory, recall or external retrieval.
• Filings consulted: 10-K filed 2026-02-11 (fiscal 2025); 10-Q filed 2026-04-22 (Q1 2026); 10-Q filed 2026-07-23 (Q2 2026 — the primary source); 8-K filed 2026-07-28 (talc resolution); 8-K filed 2026-07-29 (Firefly Bio completion, Sail Biomedicines collaboration, 2026 guidance update); 8-K filed 2026-08-04 — the day of this dive — Item 5.02, retirement of the EVP, Worldwide Chairman, Innovative Medicine effective 2026-09-01. Targeted searches run across the 10-K and both 10-Qs covered: talc, contingencies, accrual, ovarian, mesothelioma, Red River, Pecos River, Imerys; Stelara, Simponi, Opsumit, Erleada, Zytiga, Imbruvica, Carvykti, Darzalex, Tremfya, Caplyta, Xarelto; segment and therapeutic-area revenue; segment income before tax; notes payable and long-term debt, aggregate maturities, commercial paper, credit facility; operating lease and finance lease; repurchase and authorisation; dividends and consecutive-year language; Kenvue; Inflation Reduction Act and Selected Drug list; acquired in-process research and development; restructuring; the Orthopaedics separation; and risk factors.
• Filing-versus-vendor overrides (three, all resolved in favour of the filing): (1) gross margin — the vendor reports fiscal 2025 gross margin of 72.8%; the 10-K states "Cost of products sold increased as a percent to sales", and the $4.621B reconciliation gap identifies $4.6B of intangible amortisation reclassified below the gross-profit line, giving a like-for-like 67.9%, down from 69.1% — we use 67.9%, and reject the Q1/Q2 2026 quarterly gross margins on the same basis (Q2 2026 restated to approximately 68.0%); (2) net debt — the vendor reports $28.224B, the 10-K states "net debt ... was $27.8 billion", we use $27.8B; (3) the three 8-Ks — the $5.5B talc commitment, the disclosed $0.64/$1.36 adjusted-EPS reductions for 2026/2027, and today's Item 5.02 leadership change appear in no vendor field, and the second of these may invalidate the FY2027 consensus EPS this dive's multiples are built on.
• Other data rejections (detailed in Section 9): all est EBITDA, EBIT and net-income rows rejected as arithmetically impossible (net income exceeds EBIT in every forward year; FY2030 net income $43.618B against EBIT $32.971B) — epsAvg used exclusively; trailing GAAP EPS of $8.63 rejected as the valuation lens in favour of adjusted $10.86, with the 8-K's own definition cited; annual GAAP net income and EPS treated as non-comparable across FY2023-FY2025 owing to the Kenvue discontinued-operations gain (~$21.8B) and talc charges/reversals (−$5.1B in FY2024, +$7.0B in FY2025), with annual EBITDA rejected for the same reason; seg_prod treated as non-comparable to inc_a before FY2023 (FY2022 segment sales $94.943B against reported revenue of $79.990B, the difference being the consumer business); closing-price 52-week extremes preferred over intraday ($267.24/$170.59 rather than $274.90/$169.65); tech moving averages preferred over quote moving averages. Lease double-count tested and found clean, confirmed by the 10-K's "Commitments under finance leases are not significant."
• Estimate coverage: FY2026 12 analysts on EPS, 16 on revenue; FY2027 12 and 17 — the primary forward anchor; FY2028 5 and 18, with an EPS range of $10.759 to $16.579 that is too wide to anchor on and is used only as a cross-check; FY2029 2 and 9; FY2030 2 and 9. No conclusion in this dive rests on the FY2029 or FY2030 rows. The est EBITDA/EBIT/net-income rows fail the standard sanity test in every forward year and are rejected entirely.
• Peer note: the vendor peer set is appropriate — Merck ($316.24B), AbbVie ($430.74B), AstraZeneca ($241.21B), Novartis ($291.32B), Amgen ($210.50B), Gilead ($167.95B), Pfizer ($144.82B), Novo Nordisk ($196.77B), Sanofi ($102.48B) — with the single exception of UnitedHealth ($370.26B), a managed-care organisation rather than a manufacturer. No peer multiples are supplied in the file, so no peer-multiple comparison is drawn.
• Fair-value caveat: the $213 / $270 / $308 anchors are multiples of FY2027 consensus adjusted EPS (18.5x on a Sail-adjusted $11.52 for the bear; 21.0x and 23.9x on the unadjusted $12.875 for base and bull), cross-checked against FY2026E of $11.684 and, cautiously, FY2028E of $14.340. Stated arithmetic, not a discounted cash flow. The base case requires approximately one turn of multiple expansion from 19.8x to 21.0x, justified by the removal of a litigation discount rather than by any change in growth; at an unchanged 19.8x, FY2027 consensus earnings imply approximately today's price.
• Timing: Q2 2026 was reported 2026-07-15 (EPS $2.90 against $2.84 estimated, revenue $25.310B against $25.062B) and the 10-Q filed 2026-07-23. The next print is 2026-10-13, 70 days away — no earnings banner applies. However, three 8-Ks were filed in the eight days before this dive (2026-07-28, 2026-07-29 and 2026-08-04), the last of them today, and that cluster is treated as the effective catalyst environment in place of an imminent earnings date.
• Accessibility note: all tables are labelled, all emphasis is carried by text rather than colour, and every figure is stated in full rather than encoded in a chart.
• Not investment advice. Independent research, educational and informational only, never personalised. No recommendation to buy, sell or hold any security is made to any person.
• Version: 2026-08-04-full.