Johnson & Johnson

Johnson & Johnson makes medicines and the equipment surgeons use. That is the whole of it now. The baby shampoo, the Band-Aids and the Tylenol left in 2023, when the consumer business was spun out as Kenvue. What remains is two companies under one roof: a pharmaceutical business it calls Innovative Medicine, which earned 64.7% of sales last quarter, and a device business called MedTech, which earned the other 35.3% — heart pumps, catheters that map and ablate irregular heartbeats, surgical staplers and sutures, hip and knee implants, contact lenses.

A ten-year-old would understand it this way: when people get cancer, psoriasis, depression or a bad heart, J&J sells the medicine, and when they need an operation, J&J sells the tools. Both needs are permanent and neither waits for the economy.

In 2026 it expects to sell more than $100 billion of these things for the first time in its 140-year history — guidance is $100.8–101.4 billion, up 7.3% at the midpoint — and to earn $11.60–11.75 a share on its adjusted measure, up 8.2%. It has raised its dividend for 64 consecutive years. It is one of only two American companies that S&P rates AAA, higher than the United States government; the other is Microsoft.

Johnson & Johnson has always been a stock you made money on by buying when it was frightening. In 1982 somebody put cyanide in Tylenol capsules in Chicago and killed seven people; the company pulled every bottle in America and the recall became the most studied act of corporate honesty in business schools. The investors who bought in that panic did very well. The same was true, more quietly, in the talc years. This analysis is about what happens when the fright ends — because it just did.

⚠️ A timing note. J&J reports its third quarter on 13 October 2026, nineteen days after this analysis. Everything here is as of the second quarter and the 24 September price.

A year ago these shares were $175.74. They are now $270.39. Nothing that dramatic happens to a 140-year-old company because it got better at making things. It happens because the market stops being afraid of something, and in J&J's case three somethings expired inside nine months.

The first of those is the one worth staring at, because it is the one a drug company has to survive every decade. This is how the pharmaceutical business changed between the second quarter of 2025 and the second quarter of 2026:

★ J&J lost $913 million a quarter of its biggest drug and grew the division by $1.18 billion anyway. Tremfya, the successor to Stelara in psoriasis and bowel disease, added $860m on its own — up 72.5%. Darzalex, the multiple myeloma drug, added $668m and is now a $4.2bn-a-quarter product. Behind them came a queue of newer drugs growing 40–70% a year: Carvykti, Spravato, Caplyta, Rybrevant, Tecvayli, Talvey. Strip Stelara out and Innovative Medicine grew 15.5%.

That is the most important fact about the company, and it is the thing a sceptic of large pharmaceutical groups should have to answer. The standard complaint is that big pharma cannot replace what it loses. Here, in one quarter, is a big pharmaceutical company replacing what it lost with room to spare — and doing most of it with drugs it developed or had already bought years ago, rather than with a panic acquisition.

But look at the second panel before you celebrate. The concentration has moved rather than vanished. Darzalex alone is now 16.6% of the entire company's sales, and oncology as a whole is 29.3%. J&J has swapped one very large drug for another, and every drug has a patent with a date on it. The cliff is not a thing that happened in 2025. It is a thing that happens to a drug company permanently, which is the subject of Part V.

★ And the second point is about price. A year ago the market was charging you for all three fears. Today it is charging you for none of them. At $175 you were paid to take the risk; at $270 you are paying for its absence. The company is plainly better than it was a year ago. That is not the same thing as a better purchase, and most of this analysis is about the difference.

Second quarter of 2026, reported sales of $25.31bn, up 6.6% (5.6% before currency). ⚠️ Our feed's segment data stops at 2023, so these figures come from J&J's own release.

The engine is the pharmaceutical business and the passenger is the device business. Innovative Medicine grew 9.4% in the first half; MedTech grew 6.0%, and slower than that before currency. Devices earn thinner margins, grow more slowly and require more capital. They also do something useful: they are not exposed to patent cliffs, and a hospital that has trained its surgeons on a stapling system or an ablation catheter does not change suppliers lightly.

The device business is also about to get smaller on purpose. In October 2025 J&J announced it would separate its orthopaedics unit — about $9.2bn of hips, knees, trauma and spine in 2025, 9.6% of company sales last quarter — as a standalone company called DePuy Synthes, within 18 to 24 months. In September 2026 press reports said J&J was also weighing a sale, with Apollo named as a possible buyer at around $20bn. ⚠️ Neither the price nor the route is confirmed. Either way, what remains will be faster-growing and higher-margin, and the company will be that much more a pharmaceutical business.

J&J reports two sets of earnings, and they are far apart. In the second quarter its GAAP net income was $5.53bn, or $2.27 a share. Its adjusted net income was $7.08bn, or $2.90. The company wants you to use the second number, and so does nearly every analyst. The difference is mostly the amortisation of intangible assets — the accounting charge that spreads the cost of a purchased drug, patent or technology over its expected life — plus "special items" such as litigation and restructuring.

We have written about this twice before, and on both occasions we were broadly sympathetic to adding it back. At Pfizer we pointed out that amortisation is a non-cash charge that never touches the cash flow. At Thermo Fisher we quoted Buffett himself, who told Berkshire's shareholders that only about a fifth of its own amortisation charges were real costs.

★ At a drug company, we think the more honest answer is the opposite one. Buffett's point was about assets that do not wear out — a customer relationship, a brand like See's Candies — where the accounting forces a charge against something that is not actually shrinking. A drug patent is not like that. It has an expiry date written into law. When Stelara's protection ended, its sales fell by more than half in a year. Amortisation at a pharmaceutical company is not an accountant's fiction. It is the depreciation of the only asset that matters.

That changes how you should read the cash flow too. Free cash flow at J&J already pays for the research — the $14.7bn is expensed. What it does not include is the drugs the company buys to refill the shelf. Over the last six years:

Not all of that $57.4bn was spent standing still. Shockwave is growing 14.6% and Caplyta 70.9%; those were purchases of growth, not replacements. But some large part of it was the cost of refilling a shelf that empties by law, and that is why we show J&J's earnings three ways:

The dividend is safe. What is not free is the future — J&J has to buy it. An owner's true earnings here sit somewhere between the 1.5% floor and the 4.3% the adjusted figure implies, and nearer the middle than the top. That is the honest reason we do not share the market's enthusiasm at this price, and it has nothing to do with doubting the company.

For the better part of a decade the single largest question about Johnson & Johnson was not a drug. It was baby powder. Tens of thousands of women claimed that asbestos-contaminated talc in J&J's powder caused their ovarian cancer or mesothelioma; juries awarded some of them enormous sums; and J&J spent years trying to resolve all of it at once by putting the liability into a subsidiary and taking that subsidiary through bankruptcy.

★ J&J offered around $9–10 billion through bankruptcy in 2024, lost, went back to court, won on the science, and settled for $5.5 billion. Whatever one thinks of the bankruptcy manoeuvres — and three courts thought poorly of them — the end of this story is a management that took a genuine risk on the merits of its case and was vindicated by a federal judge. That deserves credit in the management score, and it gets some.

And it deserves a sense of scale. $5.5bn, spread from 2027, is about a quarter of one year's free cash flow. The first payment of up to $3bn is less than J&J spent on buybacks in 2025. This was never a threat to the company's solvency — it was a threat to its reputation and to its valuation, and the valuation has now fully recovered.

⚠️ What could still go wrong. The deal needs 95% participation, and some claimants' firms opposed earlier settlements fiercely. A shortfall would not reopen the science, which is the plaintiffs' problem, but it would keep a residue of cases alive. We will know in the coming months.

Integrity: no veto. The talc bankruptcies were aggressive and three courts rejected them, which cost a notch; the eventual result was won honestly, on the evidence. Operations: excellent — the Stelara transition is as well managed as any patent cliff we have studied. Capital allocation is the weaker leg:

⚠️ And two smaller cautions: the 2028–2030 earnings estimates in our feed come from a single analyst, so we show them in Part XI but do not lean on them; and the Graham number of $82 is meaningless for a company whose assets are patents and know-how rather than buildings.

On 14 April 2026 the board raised the quarterly dividend 3.1%, from $1.30 to $1.34 — the 64th consecutive annual increase. The annual rate is $5.36, a yield of 1.98% at $270.39. Five years ago, at the end of 2021, the dividend yielded about 2.6%. The yield has fallen because the price rose faster than the payout, and that is the first thing an income investor should notice.

★ The J&J dividend is about as safe as a dividend can be. It is also growing at 3% while the business grows at 7–8%, and it yields less than 2%. That combination is the opposite of the pattern we found at Rexford and Zoetis, where a slowing dividend preceded bad news the board could see. Here the slowing dividend comes with raised guidance. The board is not worried; it is shopping. Anyone buying J&J primarily for income today is buying a 2% yield with a 3% raise, and should compare it with the PepsiCo or Coca-Cola alternatives on the board before paying $270 for it.

So what does 23 times assume? That sales grow around 7% and adjusted earnings around 8% for years; that Darzalex's eventual decline is replaced as smoothly as Stelara's; that talc closes on the agreed terms; that American drug pricing tightens gradually rather than suddenly; and that investors keep paying a premium multiple for a AAA balance sheet and a 64-year dividend. Every one of those is plausible. Together they are a description of things going right, with no allowance for anything going wrong.

★ The arithmetic. A 2% yield plus 7–8% earnings growth gives you nine to ten per cent a year — if the multiple holds at 23. If it drifts back to the 18 or 19 times J&J has averaged over long periods, a decade of good execution returns something closer to seven. That is a respectable return from an excellent company. It is not what the investor who bought at $175 was paid for the same company a year ago, and the only thing that changed in between was the fear.

★ And the week this analysis was written, gravity came back. Buffett has long said that interest rates act on the price of every asset the way gravity acts on an apple. On 16 September the Fed raised rates for the first time in three years; on 23 September the ten-year Treasury closed above 5% for the first time since 2007. A risk-free 5.1% for a decade sits against J&J's 4.3% adjusted earnings yield and 1.98% dividend. The shares have barely noticed — defensive companies often do not, at first. But a premium multiple on a slow-growing company is precisely the thing a 5% bond yield erodes over time, and it is the most concrete way the next scare could arrive without anything going wrong at J&J at all.

We compared Eli Lilly in July and found a company worth a premium because it was growing at a rate J&J cannot match. We found Pfizer in August cheap because nobody believed in it. J&J at $270 sits between them: a company everybody now believes in, priced accordingly.