
The business, in plain English
Heart devices, glucose sensors, lab tests, baby formula and medicines for emerging markets — a Dividend King that stumbled
Abbott makes things doctors, hospitals and patients use every day. The sensor on a diabetic's arm that reads blood sugar without a finger-prick (FreeStyle Libre); the catheter that ablates a misfiring heart and the tiny pacemaker that sits inside it; the analysers in hospital laboratories and the rapid tests in pharmacies; Similac for babies and Ensure for the elderly; and branded generic medicines sold in India, Latin America and Asia. A ten-year-old's version: Abbott makes the machines, tests and foods that keep people out of hospital — or get them out faster.
It has done it for a long time. Wallace Abbott started making pills in his Chicago flat in 1888; the company has paid a dividend every quarter since 1924 and raised it for 54 consecutive years. In the quarter to June 2026 it sold $12.6bn, up 13% — but only 4.8% on a comparable basis, once the $21bn acquisition of the cancer-test maker Exact Sciences is set aside.
That gap is why the shares closed at $96.69 on 1 October 2026, 28% below their October 2025 high. In twelve months Abbott misjudged the price of baby formula and lost customers, watched its Libre sensor slow while its rival sped up, recalled three million sensors, absorbed a collapse in Chinese lab sales, borrowed $20bn for Exact, and settled over a billion dollars of litigation in four weeks. This report asks which of those things are temporary — and whether $97 is a fair price for what remains.
| Business | Q2 2026 sales | Comparable growth | In one line |
|---|---|---|---|
| Medical devices | $5.85bn | +8.4% | Libre, heart rhythm and electrophysiology, structural heart, vascular. The engine. |
| Diagnostics | $3.09bn | +2.9% | Hospital labs, rapid tests — and now Cologuard, via Exact. |
| Nutrition | $2.14bn | −3.6% | Similac, Ensure, PediaSure. Where the pricing mistake happened. |
| Established pharmaceuticals | $1.50bn | +8.7% | Branded generics in emerging markets. Quietly excellent. |
The history — a pill-maker that became a device company
Similac, a spin-off, two big acquisitions, a pandemic windfall and a formula crisis
| When | What happened | Why it matters now |
|---|---|---|
| 1888 – 1964 | Dr Wallace Abbott's "dosimetric granules" in Chicago; the 1964 merger with Ross Laboratories brings Similac and the nutrition business. | The brand at the centre of today's litigation is sixty years in the family. |
| 2004 | Buys TheraSense — the seed of FreeStyle, which becomes Libre. | The best acquisition in its history began small. |
| 1 Jan 2013 | Spins off AbbVie, its research-pharma arm (Humira). Abbott's own quarterly dividend is reset to $0.14; the streak survives only by counting AbbVie's. | The "Dividend King" record has an asterisk. |
| 2017 | Buys St. Jude Medical (~$25bn) and Alere (~$5.3bn, renegotiated down). | The heart-device franchise that is now its strongest business. |
| Mar 2020 | Robert Ford succeeds Miles White as chief executive — the 13th, all from inside. | Continuity of culture. |
| 2020 – 2022 | COVID tests, led by the $5 BinaxNOW card: $8.4bn of sales at the 2022 peak. In February 2022 the Sturgis formula plant is shut after a recall, causing a national shortage. | A windfall and a scar, in the same two years. |
| 2023 – 2025 | COVID testing falls to $297m (2025). Libre reaches $7.6bn of sales in 2025. Shares peak at $134.59 on 3 October 2025. | The base the decline started from. |
| Nov 2025 – Mar 2026 | Agrees to buy Exact Sciences (Cologuard) at $105 a share, ~$23bn including debt, funded with $20bn of new borrowing; closes 23 March. On 22 January Q4 results miss and the shares fall 10%, their worst day in years. | The biggest bet since St. Jude, made in a bad year. |
| Feb – Sep 2026 | Class I recall of Libre 3 sensors (FDA, 4 February). Low of $82.56 on 11 May. Q2 beats (+10.7% on 16 July). About $1.1bn of settlements in four weeks (August–September). $96.69 on 1 October. | The bill for old mistakes, arriving at once. |
Two lessons for an owner. The first is that the last five years flatter nobody: a shareholder who bought in October 2021 has lost about 3% a year including dividends, because Abbott's earnings were swollen by COVID tests and then had to be rebuilt without them. Measured over ten years the record is respectable — about 12% a year — and over twenty, 10%. The second is that Abbott's best acquisitions were the ones nobody noticed (TheraSense) and its most expensive ones arrived with fanfare. Exact Sciences is in the second category; whether it joins the first is the decade's test.
The circle of competence
Four businesses we understand — and one sensor war we must watch
- That Nutrition's decline, the China lab slump and the COVID cliff are behind it, and that comparable growth returns to the 6–8% range it guided for 2026.
- That Libre keeps the leadership in glucose sensors — the largest single growth engine — even as Dexcom grows faster for now.
- That Exact Sciences earns back a $23bn price and $20bn of debt, and the formula litigation is settled for sums the balance sheet can carry.
How it makes money
Devices are nearly half the sales and most of the growth
Where the money is made. The shape of the profits is plain: devices and diagnostics carry the margins, nutrition is a large but lower-margin consumer business, and established pharma is a steady emerging-markets compounder. Two features matter for an owner. Libre is a razor-and-blade business — each user buys a new sensor every couple of weeks, for years — which is why its slowdown worries the market more than any other line. And the lab business is an installed base: $6.2bn of contracted future diagnostics revenue at 30 June, from analysers that sell reagents for a decade.
The 2026 arithmetic. Management guides comparable growth of 6.5–7.5% for the year. The first half delivered 4.3%. To get there, the second half must run at roughly 9–10% — helped by easier comparisons in Nutrition and China and by new products (the Volt ablation catheter, Libre's newer sensors). The third-quarter report on 21 October is the first test.
The moat
Several moats of different widths — the weakest broke first
| The claim | The evidence | Width · trend |
|---|---|---|
| Cost and scale in glucose sensors | Libre sold $7.6bn in 2025, ~1.6× Dexcom, on a low-cost sensor; ~10–12m users of a 70–80m addressable market. But growth fell from 17% to ~9% while Dexcom grew 13%, and a Class I recall dented trust. | Wide · narrowing at the margin |
| Heart devices and hospital relationships | Electrophysiology +13% while Boston Scientific's share of US pulsed-field ablation spending fell from 100% (2023) to ~41% (early 2026); leadless pacing, mapping, LAA closure. | Wide · strengthening |
| Installed base in diagnostics | Analysers locked in for years; $6.2bn of contracted diagnostics revenue. China's volume-based procurement showed a government buyer can still reprice it. | Durable · stable |
| Brands in nutrition | Similac, Ensure, PediaSure. In late 2025 price rises drove US paediatric volumes down ~18%; Ford: "This path is not sustainable long-term." | Weaker than believed |
| Emerging-markets branded generics | +8.7% comparable; local brands and distribution. | Steady |
Buffett's 2007 letter asks for "an enduring 'moat' that protects excellent returns on invested capital", and warns that the criterion of "enduring" "causes us to rule out companies in industries prone to rapid and continuous change." Abbott is a portfolio of moats, and 2025–26 tested them in turn. The consumer brand turned out to have less pricing power than its owners believed; the glucose sensor met a rival that out-grew it; the heart business won share in the fastest technology shift in its field. On balance the moat is intact, and narrower in places than the share price assumed a year ago. We score it 7.
★★ The central question — what broke, and is it temporary?
Five problems at once. Three are passing; one is a fight; one is permanent
| What broke | The evidence | Temporary? |
|---|---|---|
| Nutrition pricing | After the 2022 shortage Abbott raised formula prices; in Q4 2025 Nutrition fell 9.1% and US paediatric ~18%. Prices were cut; the decline narrowed to −7.7% (Q1) and −3.6% (Q2). | Mostly — self-inflicted and being reversed |
| China lab sales | Volume-based procurement cut core-lab sales in China by 15–30% a quarter; Ford said in April they had stabilised. | Mostly — now in the base |
| The COVID cliff | Testing fell from $8.4bn (2022) to $297m (2025). | Over |
| The glucose sensor | Libre growth +17.4% (2025) → +9.5% (Q2 2026), against Ford's January hope of "low teens"; Dexcom +13%; a Class I recall of ~3m sensors with seven deaths outside the US. | A fight, not a fade |
| Exact Sciences and the debt | ~$23bn for Cologuard and cancer tests, funded with $20bn of borrowing; $0.20 of EPS dilution in 2026; Q2 interest $351m against $121m a year earlier. | Permanent — a choice, not a setback |
Our reading. Three of the five problems — formula pricing, China and COVID — are the kind that pass, and the evidence says they are passing: Nutrition's decline has halved twice, China has stopped falling, and COVID has nowhere left to fall. One, the sensor, is a genuine competitive fight that Abbott may or may not win; Libre is still the larger product and has new ones coming, but for two quarters its rival has been the better business. And one, Exact, is a permanent decision whose merit only time will show. Buffett's warning about expensive acquisitions in his 1981 letter fits it uncomfortably well: "If investors instead bankroll princesses who wish to pay double for the right to kiss the toad, those kisses had better pack some real dynamite. We've observed many kisses but very few miracles."
The test is dated. To meet its 2026 guidance Abbott needs comparable growth of roughly 9–10% in the second half. If the third quarter (21 October) shows the acceleration, the temporary problems were temporary. If not, the market will conclude that the base business has slowed, and the price will follow.
The competition
Winning in heart devices, losing ground in sensors, fighting on price in formula
| Competitor | Arena | Where Abbott stands | Threat |
|---|---|---|---|
| Dexcom | Glucose sensors | Q2 2026 revenue $1.31bn, +13%; G7 15-day sensor launched in the US in Q1. Out-grew Libre in the first half. | High |
| Medtronic | Sensors, ablation, neuromodulation | ~48% of US pulsed-field-ablation spending in early 2026 (Qsight estimates); Simplera sensor. | Medium |
| Boston Scientific | Ablation, LAA closure | Its US PFA share fell from 100% to ~41%; it cut 2026 guidance in April. Abbott, a late entrant (Volt approved December 2025), grew electrophysiology 13%. | Receding |
| Edwards Lifesciences | Structural heart | Abbott's US structural heart fell 0.9% in Q2. | Medium |
| Roche, Danaher, Siemens | Lab diagnostics | The same Chinese procurement squeeze hit the whole industry. | Medium |
| Reckitt (Mead Johnson), Nestlé, start-ups | Infant formula | Mead Johnson won the first federal NEC jury trial (31 Aug 2026); new entrants such as ByHeart and Bobbie. | Medium |
The pattern. Abbott's competitive position is strongest where technology moves fastest — heart rhythm, where it came late to pulsed-field ablation and still out-grew the leader — and weakest where it relied on a brand to hold prices, in formula. In glucose sensors the next two years decide whether Libre stays the low-cost leader or becomes the cheaper alternative to a better product. Unlike most of the companies on our board this month, Abbott does not face a technology that could make it unnecessary; it faces ordinary, vigorous competition in each of four businesses, which is exactly why owning all four is steadier than owning any one.
Management, ownership & capital allocation
A steady internal CEO, a bold acquisition, and buybacks at the low
| Decision | When | Our read |
|---|---|---|
| Buy Exact Sciences | Nov 2025 – Mar 2026 | ~$23bn including debt, $20bn borrowed; debt from ~$13bn to ~$33bn; ~2.7× gross debt to EBITDA. Cologuard grows mid-teens; the price leaves little room for error. |
| Buy back stock at the low | Q2 2026 | 11.6m shares at an average of $89.70 (May: $84.17) — well below our estimate of value. Exactly right. |
| Raise the dividend again | Dec 2025 | +6.8% to $0.63 a quarter, the 54th straight annual rise — while borrowing $20bn. Affordable (Part X), and a signal of confidence. |
| Settle the litigation | Aug – Sep 2026 | $670m for ~2,000 formula (NEC) claims, $385m for Sturgis under the False Claims Act, and a reported $88.5m securities settlement: ~$1.1bn in four weeks. Paying to shrink uncertainty. |
Integrity: no veto — Ford named the pricing mistake in plain words on the January call. Capital allocation: mostly sensible, with one large bet. Buying back shares at $84–90 and settling litigation are what owners would do; the Exact deal is the kind of large, debt-funded acquisition that usually disappoints, and it was made at a moment when the core business needed attention. Owner mentality: mixed signals. The chief executive bought after the crash and sold after the recovery; both trades were reasonable for him, and the second one, made a week before the shares fell 17%, is not one a shareholder can ignore. We score management and capital 6.
The numbers
Adjusted earnings, GAAP earnings and cash — and a feed that got the quarter's capex backwards
| Metric | Value | Read |
|---|---|---|
| Sales — FY2025 · Q2 2026 | $44.3bn · $12.6bn | ◆ +5.7% in 2025; Q2 +13% reported, +4.8% comparable |
| Adjusted EPS — 2025 · 2026 guide | $5.15 · $5.45–5.60 | ▲ +6–9%, after $0.20 of Exact dilution |
| GAAP EPS — trailing twelve months | $3.11 | ◆ Amortisation of acquisitions and deal costs |
| Free cash flow — FY2025 · TTM | $7.4bn · $7.8bn | ▲ ~18% of sales |
| Debt · cash (30 Jun 2026) | $32.6bn · $5.6bn | ▼ From $15.1bn of debt in December |
| Interest expense — Q2 2026 · Q2 2025 | $351m · $121m | ▼ The cost of Exact |
| Goodwill · intangibles | $35.2bn · $17.2bn | ◆ Up $23bn with Exact |
| Return on invested capital (TTM, feed) | ~6% | ◆ Depressed by goodwill; Buffett would look at returns on tangible capital |
Two kinds of earnings. Abbott's adjusted earnings ($5.45–5.60 a share guided for 2026) leave out the amortisation of past acquisitions — St. Jude, Alere, now Exact — which runs at about $2.2bn a year and has no cash cost today. That adjustment is fair, as far as it goes: the cash was spent years ago. But it means the adjusted figure never reflects what the acquisitions cost, and the GAAP figure ($3.11 over the last twelve months) is what they cost spread out. An owner should hold both in mind, and lean on free cash flow, which is roughly midway: about $4.50 a share over the last twelve months, ~$4.10 after stock-based pay.
| What the feed says | Value | What is true |
|---|---|---|
| Q2 2026 capital spending | +$399m | Shown as a positive number, so free cash flow ($2.89bn) is overstated by about $0.9bn. The 10-Q: $896m of capex in H1, i.e. ~$497m in Q2 and Q2 free cash flow of ~$2.0bn. |
| Payout ratio | 79% | On GAAP earnings depressed by amortisation. On adjusted 2026 EPS the payout is ~46%; on free cash flow, ~55%. |
| 52-week low | $100.49 | The shares closed at $82.56 on 11 May 2026; today's $96.69 is 17% above the true low, not below the feed's. |
| Product segments | $64.6bn (FY2025) | Several overlapping sets added together; FY2025 sales were $44.3bn. The first set (devices, diagnostics, nutrition, pharma) is correct. |
★ The dividend — 54 years, and the six tests
Safe, slower-growing, and carrying a new debt load
| Test | Value | Reading |
|---|---|---|
| 1 · Cover on free cash flow | 1.8× | Twelve months to June: $7.8bn of free cash flow against $4.3bn of dividends paid (1.84×); 1.78× on the new $2.52 rate. |
| 2 · The trend of the cover | 2.7× → 1.4× → 1.8× | Inflated by COVID tests in 2020–22 (2.2–2.7×), a low of 1.42× in 2023, recovering since. Exact's interest will take about $1bn a year from the numerator. |
| 3 · Funded by operations or by paper? | operations — just | Dividends are covered by cash; but in H1 2026 dividends ($2.2bn) plus buybacks ($1.2bn) exceeded free cash flow ($2.9bn), with the gap bridged by the balance sheet. |
| 4 · Balance-sheet room | ~2.7× debt/EBITDA | $32.6bn of debt, $5.6bn of cash; interest $351m a quarter. Manageable for a business of this stability, but no longer a fortress. |
| 5 · What would force a cut | nothing visible | Free cash flow would have to fall ~45%, to below $4.4bn. The formula litigation is the tail risk: ~12,700 infant claims remain after a $670m settlement covering ~2,000 — a large but one-off bill the balance sheet could absorb without touching the dividend. |
| 6 · The growth rate | slowing gently | +4.4% (2022), +8.5%, +7.8%, +7.3%, +6.8% (2026). The next rise, in December, will show whether Exact's debt has slowed it further. |
The verdict on the dividend: safe, and growing more slowly. At 46% of adjusted earnings and covered 1.8 times by free cash flow, it is in no danger. The direction is the thing to watch: cover fell in the post-COVID reset and has only partly recovered, and the $20bn of new debt will absorb cash that might have gone into faster raises. The comparison on our board is Johnson & Johnson, whose raise slowed to 3.1% as it kept cash for deals — Abbott's 6.8% is still twice that. And one honest footnote: the 54-year streak survives 2013 only because the AbbVie dividend counts. Abbott's own payout was reset to $0.14 a quarter that year.
★ Valuation — roughly fair, with a dividend while you wait
17.5 times adjusted earnings, about 24 times owner earnings
| Measure | Value | Reading |
|---|---|---|
| Price · market value (1 Oct 2026) | $96.69 · ~$167bn | 28% below the October 2025 high; 17% above the May low. |
| P/E — 2026 adjusted guide · trailing GAAP | 17.5× · 31× | On $5.53 (midpoint) and $3.11. |
| Forward P/E — FY2027 · FY2028 | 16.0× · 14.4× | Consensus $6.06 (18 analysts) and $6.72 (12). |
| Forward P/E — FY2029 · FY2030 | 13.0× · 11.7× | $7.45 (9) and $8.28 (5). |
| Free cash flow yield · dividend yield | 4.7% · 2.6% | Against a 5% ten-year Treasury. |
What is it worth? We discounted owner earnings of about $3.80–4.10 a share at 10% for ten years. If the problems of 2025–26 prove temporary and earnings grow ~9% a year for five years and 6% thereafter, ending at 18 times, Abbott is worth about $91. If Libre keeps losing ground and Exact disappoints — 5% then 4%, 15 times — about $64. If the second-half acceleration arrives and Exact works — 11% then 7%, 20 times — about $115. At $96.69 the price sits just above our central case, with a 2.6% dividend paid while you wait.
For a business we score 7 on quality, our rules ask for a margin of safety rather than permit buying at a fair price, and there is little margin here. So: hold if you own it — the dividend is safe and the problems are mostly passing — and add near $80, about 12% under our central value and a little below the May low. A strong third quarter on 21 October would lift our estimates; a weak one would move the shares toward that price on its own.
Risks, lawsuits & controversies
Verified 2 October 2026 — the formula litigation, the sensor recall and the debt
The risk we rank first is the formula litigation. Thousands of families allege that cow's-milk-based formula for premature babies (Similac and rivals) caused necrotising enterocolitis (NEC), a deadly bowel disease. Abbott won summary judgment in three federal test cases in 2025 and Mead Johnson won the first federal jury trial on 31 August 2026; but a Missouri appeals court on 5 May 2026 upheld a $495m verdict against Abbott, and a Chicago jury awarded $70m on 9 April. On 20 August Abbott agreed to pay about $670m to settle the Missouri case and roughly 2,000 other claims. About 1,700 lawsuits covering ~12,700 infants remain, and Abbott had recorded no reserve for them at 30 June. Scaled crudely from the August settlement, the rest could cost several billion dollars — large, but a one-off that the balance sheet could carry.
Second, the sensor. In February the FDA classified the recall of about three million Libre 3 and 3 Plus sensors, which gave falsely low readings, as Class I; seven deaths were reported outside the US. In medical devices, trust is part of the moat.
Third, the old formula plant. On 16 September 2026 Abbott agreed to pay about $385m to settle False Claims Act allegations over the Sturgis plant whose 2022 shutdown caused the national shortage; it remains under a five-year FDA consent decree from 2022, and the status of the Justice Department's criminal investigation is not disclosed. A securities class action over Sturgis was reported settled for $88.5m (Law360, 25 August 2026).
| Matter | Status on 2 October 2026 | What is at stake |
|---|---|---|
| NEC preterm-formula litigation | $495m Gill verdict affirmed (5 May); $70m Chicago verdict (9 Apr); ~$670m settlement of Gill + ~2,000 claims (20 Aug); ~1,700 suits / ~12,700 infants remain; federal MDL test cases mostly won by defendants. | Possibly several billion dollars over years; no reserve recorded at 30 June. |
| Sturgis plant | ~$385m civil False Claims Act settlement (16 Sep 2026); FDA consent decree since 2022; DOJ criminal probe status not disclosed. | Largely crystallised. |
| Securities class action (Sturgis) | Reported settled for $88.5m (Law360, 25 Aug 2026); derivative suit settled (final approval June 2026). | Closed. |
| Libre 3 / 3 Plus recall | FDA Class I (4 Feb 2026); ~3m sensors; 7 deaths and ~860 serious injuries reported outside the US. | Trust, share, possible product-liability claims. |
| Exact Sciences integration | Closed 23 Mar 2026; $0.20 of 2026 EPS dilution; amortisation $658m in Q2 vs $420m. | Returns on a ~$23bn purchase. |
Berkshire never owned Abbott. It did own, for a short while, the drug company Abbott spun off in 2013, which tells you something about my record of picking the right half of a family. I have always admired businesses like this one from a distance — companies that make the ordinary, necessary things of medicine and raise their dividend every year whether or not anyone is watching. This year people were watching, and not kindly.
The business deserves its due first. Abbott sells heart devices that are winning share in the fastest technology change in their field; the glucose sensor that is still the largest and cheapest in the world; lab analysers that sell reagents for a decade; and branded medicines in the countries where most of the world's new patients will come from. It has paid a dividend every quarter since 1924 and raised it fifty-four years running. Last quarter its electrophysiology business grew thirteen per cent and its emerging-markets medicines almost nine.
Now the rub. Abbott raised the price of baby formula after causing the 2022 shortage, and in late 2025 the parents walked; the chief executive said the path was not sustainable, which was honest and late. Its sensor grew half as fast this year as last, while Dexcom's grew faster, and it recalled three million of them. It borrowed twenty billion dollars to buy Exact Sciences — a good business at a full price — just when its own needed attention; I wrote in 1981 that we have observed many kisses but very few miracles. And it is paying for old mistakes: more than a billion dollars of settlements in four weeks, with some twelve thousand infant claims still outstanding.
At ninety-seven dollars you pay seventeen and a half times this year's adjusted earnings and about twenty-four times what I would count as owner earnings, after the stock pay and the new interest bill. On our central reckoning the business is worth about ninety-one; if its problems prove temporary and the second half accelerates as management promises, nearer a hundred and fifteen. The dividend, at two and a half per cent, is covered almost twice by free cash flow and is in no danger.
So hold it if you own it, and add near eighty dollars. That is a little below where the shares sat in May, and about twelve per cent under our central value — the margin a business of this quality, in the middle of a repair, should offer a new buyer. The test comes on the twenty-first of October: if comparable growth reaches the high single digits, the passing problems were passing, and I will raise my numbers. If it does not, the market will very likely offer the eighty dollars on its own.
— The Buffett Lens · Dividend Line Research · from an admirer at a distance, still at a distance, but closer than last year
Add Near $80Five problems at once: formula pricing, China and the COVID cliff are passing; the Libre-versus-Dexcom fight is open; the debt-funded Exact deal is permanent. ★ At $96.69 — 17.5× adjusted earnings, ~24× owner earnings — the price sits just above our central value (~$91), with a safe 2.6% dividend covered 1.8× by free cash flow. Hold; add near $80. Q3 results on 21 October: watch for high-single-digit comparable growth.

