Novo Nordisk

IThe business

Novo Nordisk discovers, makes and sells medicines for diabetes, obesity and rare diseases. It is run from Bagsværd, outside Copenhagen, it employed 69,505 people at the end of 2025, down from 77,349 a year earlier after it cut about 9,000 positions, and it reports in Danish kroner.⁵ Sales were DKK 111.8B in 2016 and DKK 140.8B in 2021. They were DKK 309.1B in 2025 C (calculated).¹⁰¹²⁵ I convert to dollars at DKK 6.6895 to the dollar, the ECB rate at the Q4 2026 close, and say so each time C (calculated).³

In 2025 sales rose 6.4% in kroner and 10.3% at constant exchange rates, operating profit was DKK 127.7B, a margin of 41.3%, and net profit DKK 102.4B C (calculated).⁵ Three therapy areas make it up. Diabetes care was DKK 207.1B, obesity care DKK 82.3B, and rare disease, which means bleeding disorders and growth hormone, about DKK 19.6B C (calculated).⁵ The company counts 45.6 million patients treated.⁵

What it sells

• Obesity. Wegovy, the weekly semaglutide injection at 2.4 mg, now also at 7.2 mg as Wegovy HD in the US since April 2026, and the Wegovy pill, a daily 25 mg tablet launched in the US on 5 January 2026. Saxenda, the older daily injection, is fading. Total obesity care was DKK 82.3B in 2025.⁵⁶
• Diabetes. Ozempic, the weekly injection, was DKK 127.1B in 2025, the largest product the company has. Rybelsus, now rebranded Ozempic pill in the US, was DKK 22.1B. Then the insulins, Tresiba, NovoRapid and Fiasp, where biosimilars exist and sales fell 7% in the first half at constant rates.⁵⁶
• Rare disease. NovoSeven and Alhemo for haemophilia, Norditropin and Sogroya for growth hormone. The FDA has held up denecimig, a new haemophilia A antibody, because of manufacturing remediation, and gave no new date on 2 October.⁶¹⁴

The three semaglutide brands, Ozempic, Wegovy and Rybelsus, added up to DKK 228.3B, or 73.9% of sales C (calculated).⁴ Of that, DKK 148.3B was sold in the United States, 48% of the company C (calculated). One molecule is the business. That is the first thing to hold in mind for everything below, and it is the same shape as the Lilly page on this site, with tirzepatide at 64%.⁴²⁴

What a dollar of list price becomes

The prices people quote are not what Novo receives. In 2025 gross sales were DKK 729.4B before rebates, discounts, charge-backs and returns. Those took DKK 420.4B, and net sales were DKK 309.1B. Novo kept 42.4 cents of each krone of list price, against 38.2 cents in 2023 C (calculated).⁵ US managed care and Medicare alone took DKK 247.0B.⁵ The sales-deduction liability on the balance sheet at 30 June 2026 was DKK 140.2B, about 45% of a year of sales, money owed back to payers for sales already booked C (calculated).⁶

This is why list-price news matters less than it sounds, and also why it matters. From 1 January 2027 Novo cuts the US list price of Wegovy and Ozempic to $675, about 50% off Wegovy and 35% off Ozempic. The company says the change will affect 2027 cash flow.⁶ I read the cut as mostly shrinking the rebates taken from the list price. How much of it reaches net sales is the single biggest number I cannot know, and I treat it as a range below.

Where it sells

The United States was 51% of adjusted sales in the first half of 2026, DKK 76.3B, and International Operations 49%, DKK 72.3B C (calculated).⁶ The two halves are moving in opposite directions. US adjusted sales fell 4% at constant exchange rates in the half. International sales rose 8%, and international obesity care rose 40%, led by Europe and Canada up 57%.⁶ China was flat in the half, and Novo cut Wegovy list prices there in late 2025.⁶

Volume up, price down

This is the pattern of the whole business now. Novo is selling a lot more medicine for a lot less per patient. In the United States, obesity care grew 6% in the half. Behind it, Wegovy injections fell 17% on price while the Wegovy pill added DKK 5.4B in six months, and combined Wegovy sales rose 7%.⁶ The pill is the story of the year. It passed 5 million prescriptions by mid-July, six months after launch, with weekly prescriptions above 265,000 in the week to 17 July, and the company calls it the strongest US GLP-1 volume launch ever.⁶ It sells for $149 to $299 a month for cash-paying patients, depending on dose.⁶ Injectable Wegovy still fills about 310,000 prescriptions a week, about 120,000 of them paid in cash through NovoCare or telehealth partners such as Ro, WeightWatchers and LifeMD.⁶

Medicare is a new payer. Under the Bridge programme, in place since 1 July 2026 and running under the November 2025 agreement with the US government, Medicare Part D covers obesity medicines for a majority of beneficiaries.⁶ Novo calls early uptake encouraging and gives no number. Lilly gave one on its call: about 7 in 10 new Bridge patients chose a Lilly medicine.¹⁷ I put the two side by side because neither can be checked against the other, and only one of them has published a figure.

Gross margin shows the cost. The adjusted gross margin was 78.2% in the second quarter of 2026 against 82.7% a year earlier, from lower prices, about DKK 3B of one-off costs to shrink manufacturing contracts, and currency C (calculated).⁶ The adjusted operating margin rose to 42.5% from about 41.7%, helped by sales and distribution costs falling 13% after the restructuring.⁶

IIThe moat

What keeps a rival from taking the profit? I look in four places: the patent, the factory, the evidence and the customer. Here the patent is real and has a date, the factory is real and very expensive, the evidence is strong and the customer is the weakest link.

The patent

The US compound patent for semaglutide, in Ozempic, Rybelsus and Wegovy alike, ends in 2032.⁴ In China the Supreme People’s Court upheld Novo’s semaglutide compound patent in December 2025. That settles validity. It does not move the end date.⁴ In some other markets the position is already different. The semaglutide compound patent in India expired on 20 March 2026, and during the first quarter competitors launched generic semaglutide there. In Canada and Brazil none had launched yet, though the first Canadian generics were approved, and in the second quarter Novo said loss of exclusivity in Canada had cut realised prices while its volumes held.⁷⁶

In the US, generic makers have filed for approval of copies of Ozempic, Wegovy, Rybelsus, Victoza and Saxenda, and Novo has sued. It has settled the Ozempic cases with Mylan, Sun, Dr. Reddy’s, Apotex, Alvogen, Rio and Zydus on confidential terms. Cipla’s filing is still in court.⁴⁶ A settlement usually fixes an entry date that the public does not see. I cannot model that date, and I assume 2032 for the base with erosion starting in 2032 and 2033.

The factory

Making a peptide medicine for tens of millions of patients is hard. Novo spent DKK 60.1B on property, plant and equipment in 2025, 19.5% of sales, against DKK 5.8B in 2020 C (calculated).¹²⁵ The 2024 acquisition of three former Catalent manufacturing sites sits inside DKK 82.2B of cash paid for businesses that year.¹²⁵ Net property, plant and equipment was DKK 231.8B at 30 June 2026, up from DKK 208.4B at the end of 2025.⁶ Guidance is about DKK 55B of capex in 2026, and the company says it expects capex to decline in the coming years.⁶

That is a real barrier, and it is also a bill. Gross margin has already fallen from 84.7% in 2024 to 81.0% in 2025, partly from the depreciation of the Catalent sites, and in the first half of 2026 the adjusted margin was 79.3%.⁵⁶ A factory protects profit only if it is full. In the second quarter of 2026 Novo took about DKK 3B of one-off costs to right-size manufacturing capacity agreements.⁶ The factory was built for a growth path that moved.

The evidence

Outcomes data on the label is a barrier a new entrant takes years to match. The EU label for the Wegovy pill includes the SELECT result, a reduction in major cardiovascular events.⁶ Oral semaglutide cut major cardiovascular events by 14% in the SOUL trial, and Japan and China have approved the label update.⁶ Wegovy is also the first and only GLP-1 approved in the US for MASH, a liver disease, and the FDA has accepted a filing to add MASH data to the pill with a decision due in the first quarter of 2027.⁵⁶ This is the part of the protection I trust most. It lasts past the patent, because doctors and payers still prefer the drug with the proof.

The customer, and where it leaks

Here the protection is thin. Insurers and governments buy on price, and in 2025 they took 57.6% of the gross price in deductions C (calculated).⁵ Diabetes care value share fell 3.6 points to 30.1% in a year, from losses in both operating units.⁵ US insulin sales fell 17% in the first half.⁶ Novo sued to overturn the price-setting provisions of the Inflation Reduction Act and lost. The Supreme Court declined to hear the case on 18 May 2026 and the matter is closed.⁶ A compounding pharmacy, Strive Specialities, has an antitrust suit against Novo and Lilly in Texas, alleging agreements with telehealth firms that bar them from working with compounders.⁷

Overall I would call the protection wide until 2032, and narrower than it was in 2023, when prices were higher and the rival’s pill and stronger injection did not exist. It protects the volume. It does not protect the price, and the price is where the profit is.

IIIThe competitive advantage

Novo and Eli Lilly make the two leading families of obesity and diabetes drugs, sold as injections and pills to the same patients. Lilly is the sensible comparison, and the Lilly page on this site is the fairest benchmark for how I judge Novo.²⁴ I use the same discount rate, 9.0%, the same ten-year method and the same margin-of-safety rule so the two pages can be read together.

Where Novo is ahead

• Breadth. The company says Wegovy has launched in about 60 countries and that its medicines treated 41.6 million people with diabetes and 4.9 million with obesity in the half.⁶
• Volume in branded obesity medicines. IQVIA data in the Annual Report gives Novo a 59.6% worldwide share by volume at November 2025.⁵ In the US, the company says Wegovy remains the leader among branded obesity medicines by new patient starts. Lilly says it holds about 6 in 10 of all US obesity prescriptions and more than 7 in 10 of injectables.⁶¹⁷ Both can be true. Share of new starts is one measure, share of the stock of patients another, and neither company shows the working.
• The first obesity pill. Wegovy pill launched on 5 January 2026, with 16.6% mean weight loss in its OASIS 4 trial, and has been approved in the UK, the UAE and the EU.⁶ Lilly’s pill, Foundayo, is newer. Lilly says about 4 in 10 US patients starting an oral obesity medicine chose it.¹⁷ If those are the only two pills, that leaves about 6 in 10 for the Wegovy pill E (estimate).

Where Novo is behind

• Growth. Lilly’s third-quarter revenue rose 39% to $24.4B and its 2026 guidance is $91.0B to $92.0B.¹⁵ Novo guides adjusted sales growth of 0% to −6% for 2026, at constant exchange rates, and the second quarter was +7%.⁶
• The molecule. In REIMAGINE 4, an open-label head-to-head in type 2 diabetes, CagriSema, Novo’s next injection, met non-inferiority to tirzepatide on weight, 15.2% against 15.8%, and did not on blood sugar, 1.9 points against 2.2, on the basis that everyone stayed on treatment.⁶ It is the cleanest head-to-head I know of between the two pipelines. On 28 October Lilly announced topline results for its triple agonist retatrutide against Ozempic 2 mg in type 2 diabetes: weight loss of 19.5% and 20.9% against 9.4%, and a blood sugar fall of 2.2 points against 1.8, at 80 weeks. The trial was open-label and Lilly gave the efficacy-estimand figures.¹⁸
• The pipeline setbacks. The ZEUS trial of ziltivekimab, a heart drug, missed its main goal in the first half of 2026. Novo took DKK 6.3B of pipeline impairments in the second quarter, DKK 4.0B of it for monlunabant, an obesity candidate. Those are inputs to my view of research returns.⁶

The next five years of products

What is coming from Novo is not small. CagriSema has a US decision due in the fourth quarter of 2026. The higher-dose Wegovy and the single-dose pen are rolling out. Zenagamtide, formerly called amycretin, entered phase 3 in obesity and a heart-failure outcomes trial that reads out in the second half of 2029. The next CagriSema trial in a higher dose reads out in the first half of 2028.⁶ At its Capital Markets Day on 21 September, management said it would launch at least five multi-blockbuster medicines by 2030 and grow revenue in line with other large drug makers, and that it would do more deals. STAT reported that the shares fell 5% during the opening remarks.¹⁹

The table says two things at once. Lilly is growing, taking share and generating the data. Novo’s US sales are shrinking, it is losing diabetes share and buying time with a pill. The price differences follow that. Lilly sells at 30 times next year’s earnings and Novo at 11.7 times, a 61% discount for a business that earns a similar margin.²⁰²⁴ I do not think the gap is a mistake. The market pays for growth it can see and discounts growth it cannot.

What I do take from the Lilly page is the standard it sets. At 9.0% that page finds Lilly’s price 91% above its base, and Novo’s is 43% above. Both fail my margin-of-safety test. Novo fails by less, because its price assumes much less, and I would call that a lower price for a weaker business. I give no rating, and I do not read Novo as a cheaper Lilly. The Lilly page makes a bet that one medicine class keeps growing for a decade. This page makes a different bet, that a large cash machine can stay roughly the size it is, while its main product is repriced and loses its patent.²⁴

IVManagement

Management is new, the board was replaced in November 2025, and one owner holds the votes. I take it in the order I would take it for any company: who decides, how they are paid, how much of their own money is in, and what they did with the cash over a full cycle.

Who decides

Maziar Mike Doustdar became president and CEO on 6 August 2025. He joined Novo in 1992 and had run International Operations since 2015, the part of the company that is growing now. He followed Lars Fruergaard Jørgensen, who left after more than 34 years.⁹ At an extraordinary general meeting on 14 November 2025 the chair, vice chair and the independent members stepped down. Lars Rebien Sørensen became chair, Cees de Jong vice chair, and Britt Meelby Jensen and Stephan Engels joined, which cut the board from 12 to 9.⁴⁹ The CFO is Karsten Munk Knudsen.⁹

Within a year the new team cut about 9,000 positions for DKK 8B of yearly savings, launched the first oral GLP-1 for weight loss in the US, bought a liver-disease company, Akero, which closed on 9 December 2025, and raised 2026 guidance twice.⁵⁴⁶ These are real changes, and I note them before I note anything else.

How they are really paid

Pay is what tells you what a board wants. The CEO’s 2025 pay was DKK 20.7M, fixed pay from 7 August plus a bonus for the full year, of which 31% is not yet final and 35% is deferred.⁹ The yearly bonus has four equal parts. They are sales growth, operating profit growth, a set of strategic initiatives, and an individual assessment.⁹ The 2025 outcome shows how it works. Sales grew 10.6% against a minimum of 15.1%, operating profit 6.4% against a minimum of 17.4%, so those two parts paid nothing. Ten of twelve initiatives and the individual score paid out, 24.7% of the maximum in total, which is 2.96 months of salary.⁹

The three-year share plan uses the same ideas. For the 2023 plan, average sales growth of 24% and average profit growth of 25% each hit the cap, and initiatives scored 25 of a possible 33 points, so the plan paid 92% of the maximum.⁹ The executives’ shares vested at DKK 87.6M, and share price development over the three years took DKK 35.1M off that value.⁹ In the three years the plan covers, the share price went from DKK 469 to DKK 325.⁵ In the tables I could read I find no measure of return on capital and none of the share price.⁹

What does that pay a manager to do? Grow sales, grow operating profit and finish projects. It does not charge them for the money they use. Return on invested capital was 88.5% in 2023 and 39.3% in 2025, because capital spending rose nearly tenfold from 2021 while operating profit was flat in 2025.⁵ I do not say the plans were wrong in 2023, when the company was growing 30%. I say that this pay design would not have asked whether DKK 60B a year of factories was too much. The former CEO left with DKK 42.9M of severance and DKK 22.0M for a one-year non-compete, as the contract provided, and he keeps his rights under the share plans.⁹

How much of their own money is in

The CEO must hold three times his base salary in shares and the executives 1.5 times, and the company says all met it.⁹ The CEO built his holding in 2025, adding 36,956 shares and selling 1,852. Non-registered executives sold 123,884 shares in the year and added 120,136.⁹ Against the stake of the Lilly chief executive on the Lilly page, about $1.16B, these are small.²⁴ They are real money, and they have lost about 18% of their value since the year-end.

The owner behind the owners

The Novo Nordisk Foundation owns its stake through Novo Holdings A/S. The Holdings company owns all 1,075M A shares and 177.6M B shares, 28% of the capital in nominal value, and each A share carries 100 votes against 10 for a B share. That gives it about 77% of the votes, which I calculate from the share capital note C (calculated).⁵ A holder with that many votes decides who sits on the board. The Foundation’s aim is to keep a medicine company strong for a very long time. That suits a patient owner and may differ from what a buyer who wants a quick return would choose.

What they did with the money

The record covers seven years, 2019 to 2025. Net profit added up to DKK 471.5B C (calculated). Cash from operations less capital spending and stock pay, which I call owner cash, was DKK 405.6B C (calculated).¹² Out of that came DKK 177.0B for deals, DKK 127.2B for buybacks and DKK 214.0B for dividends, DKK 518.2B in all C (calculated).¹² The gap of about DKK 113B was borrowed. The company had net cash in 2019 and net debt of DKK 95.4B at the end of 2025.¹²⁵

Buybacks are the clearest test of judgement. From 2022 to 2024 Novo spent DKK 74.2B buying its shares, when the year-end price was DKK 469, DKK 698 and DKK 624. In 2025, at DKK 325, it bought DKK 1.4B.¹²⁵ It resumed in 2026 with a DKK 15B programme, and since 6 May it has paid an average DKK 291 a share, above today’s DKK 268.¹³ The share count fell from 4,620M at the end of 2021 to 4,465M at the end of 2025.⁵ That is 155M fewer shares for DKK 75.6B, about DKK 490 each, nearly double today’s price E (estimate).¹²⁵ It paid most when the price was highest and almost stopped when it was lowest, in the year it spent DKK 30B on Akero and DKK 60B on plants. Cash was short, which explains it and does not make it good timing.

The deals are harder to judge. Akero cost about DKK 30B in 2025 and is a phase 3 liver-disease drug, so its value is not yet known.⁵¹² The Catalent sites bought capacity for a growth path that has since slowed. Impairments of intangible assets were DKK 9.5B in 2024 and DKK 2.8B in 2025, and DKK 6.3B more followed in the second quarter of 2026.¹²⁶ I cannot say any single deal was a mistake. I can say the pattern is of a company that grew with its product and spent as if the growth would not slow.

What would change my mind about management is plain. A cost line that stays down while R&D rises, capex falling, as the company says it will, a payout policy that keeps buying when the price is low, and a pay plan that charges for capital.⁶

VFinancials

I take ten years of reported figures, then the twelve months to 30 June 2026 on an adjusted basis, because the 340B reversal changes the picture so much. Everything is in Danish kroner, the currency the company reports in. Pre-2019 cash flows follow the older accounting before IFRS 16 put lease payments in financing, which moves a few billion between lines and not the story.¹⁰¹²

A margin that does not move

For ten years the operating margin stayed between 41.3% and 44.2% while revenue grew from DKK 112B to DKK 309B C (calculated).¹⁰¹²⁵ In 2025 it was 41.3%, held down by about DKK 8B of one-off restructuring costs, and without them it would have been 43.9% C (calculated).⁵ In the twelve months to June 2026, adjusted, it was 40.2% C (calculated).⁶ I take that as the level the business earns today, and my base case moves it down for two reasons, the price changes of 2027 and the depreciation of DKK 230B of plant.

The capital bill

Capital spending went from 4.5% of revenue in 2021 to 19.5% in 2025 C (calculated).¹² Free cash flow, meaning cash from operations less spending on property, plant and equipment, peaked at DKK 83.1B in 2023 and fell to DKK 59.0B in 2025 C (calculated). The company’s own free cash flow for 2025 of DKK 28.3B also subtracts purchases of intangibles such as Akero, under the definition it used before 2026.⁵¹² Guidance for 2026 is DKK 45B to 55B.⁶

Return on capital

A return of 39.3% is still very high. The direction is what matters. New plant earns less than the old plant did, because it is not yet fully used.⁵

What an owner could take out

Net profit is not the number I value. An owner gets cash after the stock pay, after the money needed to keep the factories and the pipeline from decaying, and after any deals the company must do to replace what loses a patent. I call that stand-still owner cash. It cannot be known exactly, and I say so, because the main input, how much of today’s DKK 55B of capex is only maintenance, is not disclosed.

Stock pay is small here. The share-based payment expense was DKK 1.4B in 2025, 1.4% of net profit, and the company buys shares to cover it.¹² Depreciation of property, plant and equipment, with leases, was only DKK 8.1B in 2025, but it sits on a cost base that was built mainly in the last three years.¹² That makes depreciation a floor for maintenance capex, not a measure of it. Replacing DKK 232B of plant over a 25 to 30 year life costs DKK 8B to 9B a year, and equipment and tooling wear faster. I show three levels, DKK 15B, 25B and 35B, and use the middle in the reverse DCF.

At the middle level an owner takes out DKK 71B, about 70% of net profit C (calculated). That is a 6.0% yield on the market value, and it is the figure the reverse DCF below starts from. The twelve-month free cash flow of DKK 75.8B looks higher, but I do not use it, because the first half of 2026 had a DKK 12B rise in tax payable not yet paid, and the company’s own full-year guide is DKK 45B to 55B.⁶

The balance sheet

At 30 June 2026 borrowings were DKK 140.1B and cash and securities DKK 45.0B. The company’s own net debt, which leaves out leases, was DKK 86.5B, and with leases it is DKK 95.1B C (calculated).⁶ I use the second, because leases are real obligations. Equity was DKK 221.3B, 37.2% of assets. The company has EUR 15.0B and CHF 1.09B of bonds, rated Aa3 by Moody’s and AA by S&P, and it repaid a EUR 1.3B bond in May.⁶ Net debt is 0.8 times adjusted operating profit C (calculated). Debt is not the risk here. The risk is that it was taken to fund capacity for a growth path that has slowed.

What 2026 looks like

The company guides adjusted sales and adjusted operating profit each 0% to −6% at constant rates, raised on 4 August from −4% to −12%, which was itself raised in May from the −5% to −13% first given in February. On the unadjusted basis the middle of the range would be +5% for sales and +12% for operating profit, because the reported figures include the DKK 26.8B reversal.⁶⁸ In kroner the company expects adjusted sales to be about one point lower and operating profit about two points lower than at constant rates.⁶ Capital spending is expected to be about DKK 55B, the tax rate 21% to 23% and the free cash flow DKK 45B to 55B.⁶

The 340B reversal deserves a sentence. It is a US programme that obliges drug makers to sell at a discount to safety-net hospitals and clinics. Novo had set aside a provision of $4.2B for the programme, and in the first quarter of 2026 it reversed all of it, which added DKK 26.8B to reported sales and operating profit with no cash. I strip it out of every figure I use, as the company does in its adjusted measures.⁶

Earnings per ADR were DKK 23.03 in 2025, against DKK 10.37 in 2021, and the dividend rose from DKK 5.20 to DKK 11.70 with a payout ratio of 50.7%, a yield of 4.4% at today’s price C (calculated).⁵ One ADR is one B share.⁵ Shares outstanding are fewer than the 4.443B weighted average for 2025, because the buybacks continue. Total shares are 4,465M, less 44.2M the company held as treasury shares on 3 August, and about 4,420M is the figure the second-quarter earnings imply C (calculated).⁶⁵

VIValuation

What would ruin it

Before any number, I ask how an owner gets hurt here. I see three ways, and two are about the company.

1. The price of the class falls faster than the volume grows. The US list price of Wegovy and Ozempic falls to $675 on 1 January 2027, the cash-pay price of the pill is $149 to $299, and Medicare pays under a government agreement.⁶ US adjusted sales already fell 4% in the first half while prescriptions rose.⁶ The medicine stays good and the profit on it shrinks. This is the likeliest way for the business to disappoint, and it is my bear.
2. A better molecule takes the patients. 74% of sales are semaglutide C (calculated). If tirzepatide, retatrutide and Lilly’s pill keep winning the new patients, Novo is left with the older ones, at falling prices, and the factories it has just built run below capacity.⁴⁶¹⁷¹⁸ This is the one I cannot rule out, and it is why I stay cautious about any growth number above my base.
3. The owner pays too much. This needs nothing to go wrong at Novo. The business can do exactly what I expect and the owner can still earn little, because the price already pays for it. On my numbers, at $40.00, that is true: my base returns 4.7% a year against the 9.0% I ask.

The first two sit in my bear case, which is worth $12. The third is what the rest of this section measures.

What the price implies

At $40.00 the ADR is worth $176.8B, DKK 1,183B, on about 4,420M shares outstanding C (calculated).¹³⁶ Add DKK 95.1B of net debt including leases and the business is valued at DKK 1,278B C (calculated). The stock is down 73% from its closing high of $146.91 on 25 June 2024, and 21% from $50.88 at the end of 2025.¹ The Copenhagen close on 29 October was DKK 260.35, and the buyback notice of 26 October shows purchases at DKK 257 to 263 a share in the latest week, so the conversion of DKK 267.6 agrees with where the shares traded.²¹³

Start with earnings. The ADR is 11.7 times the 2027 Nasdaq consensus of $3.42, a mean of seven estimates, and 11.3 times the 2028 mean of $3.53 from three C (calculated).²⁰ Consensus has 2027 earnings flat on 2026. My own base has 2027 earnings of $3.02, 12% below the consensus mean, because of the US list-price cut and my margin E (estimate). I take the gap seriously. If I am too low, the base moves up by about $1 for each point of margin E (estimate).

Only Pfizer is cheaper on 2027, and consensus has Pfizer’s earnings falling in 2028. A low multiple is what a market pays for a business it thinks will stop growing. It is not by itself a sign of error.²⁹

The reverse DCF starts from the middle stand-still owner cash of DKK 70.7B. I solve for one growth rate for ten years, take a 10% cut in year ten for patent erosion, then grow 3%, at my 9.0% rate. The price needs 4.8% a year, which takes owner cash to DKK 113B in year ten C (calculated). Owner cash after stock pay grew 7.4% a year from 2019 to 2025, from a base that was lower, and fell 29% from its 2023 peak to 2025 C (calculated).¹²⁵ My base grows it 0.7% a year, my bull 5.9% and my bear −6.6% E (estimate). At 8.0% the price needs 2.4% and at 10.0% it needs 6.9% E (estimate).

The price asks stand-still owner cash to grow about 5% a year for ten years. My base has it growing under 1%, and my bull about 6%.

The net debt line needs a note. The second half of 2026 is expected to be weak for cash: the guide of DKK 45B to 55B for the year against DKK 55.3B made in the first half implies free cash flow near −DKK 5B in the second half, and the interim dividend of DKK 3.75 a share, about DKK 16.6B, is paid in August. The buyback programme will take about DKK 8B more.⁶ I deduct all of it from value at 1 January 2027, so I do not give credit for cash that will have left the company by then.

The discount rate

I use 9.0%, the hurdle on my other large, established companies and the rate the Lilly page uses, so the two can be compared.²⁴ The capital asset pricing model gives less. With a 4.94% ten-year Treasury and Damodaran’s 4.20% premium, Novo’s month-end betas against the S&P 500 of 0.57 over ten years, 0.69 over five and 1.30 over three give 7.3%, 7.8% and 10.4% C (calculated).²¹²²²³ The three-year figure is high because the shares fell sharply while the market rose. None of the three is a rate I would put on a business whose main product is being repriced.

Novo gives its own answer. In testing assets for impairment, it uses post-tax discount rates of 7.0%, or 8.3% before tax.⁵ That is management’s number for its own cash flows, and it sits near the bottom of the CAPM range. On it my base is $43 E (estimate). The base equals the price at 7.3% E (estimate). So the market is pricing my base cash flows at about the rate Novo uses for its assets. That is the plainest way to say what is going on. Whether you prefer 7.3% or 9.0% is most of the debate.

The base reaches the price only in the top rows. At 8.0% it is $34, 15% below the price. At 7.0% it is $43, 8% above.

What I cannot judge

Five inputs drive the value and none can be known today. I show ranges for that reason.

• What the 2027 list-price cut does to net price. The company says it will affect cash flow in 2027 and gives no figure.⁶ A 20% swing in revenue every year moves the base from $28 to $22 or $35 E (estimate).
• The date and terms of US generics. Settlements fix dates that are not public. I assume erosion from 2032.⁴
• What CagriSema, zenagamtide and the new pills are worth. CagriSema has a US decision in the fourth quarter of 2026. The others are in phase 3 or earlier.⁶ They have data and no approval, and a trial against tirzepatide did not clear the bar on blood sugar.⁶
• How much of DKK 55B of capex is maintenance. I show DKK 15B to 35B.
• The krone. Most costs are in kroner and most sales are in dollars and other currencies. A 5% move in the dollar changes adjusted operating profit by DKK 5.0B over twelve months, and the company hedges the dollar for twelve months.⁶ I take today’s rate and do not forecast one.

Three cases

I build each case from obesity revenue, diabetes and rare disease revenue, operating margin, tax, capital spending, depreciation, working capital at 20% of added revenue and deals as a share of revenue C (calculated). Stock pay is inside the margin. Deals are charged every year as a cash cost: a company that loses a patent in 2032 has to buy or build what comes next. Novo’s deals were 12.7% of revenue over 2019 to 2025, and 6.8% without the Catalent sites C (calculated).¹² I use 5% in the base and the bear, a smaller sum on smaller revenue, and 4% in the bull.

The year 2032 needs care. The US compound patent for semaglutide ends then.⁴ I do not draw a cliff. In the base, obesity revenue falls 4% in 2033 and diabetes revenue 7%, because I assume generics arrive at a lower price. The cut to year-ten cash flow is 10% in the base, 20% in the bear and none in the bull. The deals I charge are meant to pay for the replacements.

The base is a flat company that gets a little better and then worse. Total revenue goes from about DKK 294B in 2026 to DKK 354B in 2032, then back to DKK 342B in 2036 E (estimate). Obesity care grows from about DKK 95B to DKK 184B in 2032, 11.7% a year, with volume strong and price down. Diabetes falls from about DKK 199B to DKK 170B. The operating margin falls from 40% to 36%. Base earnings per ADR are $3.02 in 2027, below the $3.42 consensus mean E (estimate).²⁰

The bear has the price of the class fall faster than volume grows, Wegovy lose share to tirzepatide and its successors, generics arrive early in the US, and the margin settle at 28%. Revenue falls to DKK 218B. Novo stays a profitable company, and an owner at $40.00 still loses 4.4% a year E (estimate).

The bull has the pill and the 7.2 mg dose win the next wave of patients, CagriSema and zenagamtide approved and competitive, International Operations growing as Europe does now, and margins back at 42% to 43% as capex falls. Revenue reaches DKK 448B, 45% above 2025 C (calculated). That case is worth $49, and the price is 18% below it.

Weighted 25%, 50% and 25%, the cases give $29 a share E (estimate). At 20%, 50% and 30% the figure is $31, and at 30%, 50% and 20% it is $28 E (estimate). Against the price the skew is −70% to my bear and +22% to my bull E (estimate). A mid-year convention adds about $1 to the base E (estimate).

The Treasury pays 4.94% and my Novo base returns 4.7% at the price, so on my numbers the business matches a government bond and carries every risk above. The best comparable on this site is Prosus, whose price is 22% below its base and whose base returns 9.8%.²⁵²¹ Lilly, at 1.3%, is the worst on the list, with the price furthest from its value. I rate neither. I only say that at their prices Novo is closer to paying its way than Lilly is, and that the best return on this site is not here.

What closes the gap

Operating changes can close it, but not small ones. Revenue would need to be 37% above my base every year, DKK 467B in 2030 and DKK 470B in 2036, or margins about 11 points higher in every year E (estimate). Both are above anything in the record. The discount rate closes it alone at 7.3%. The market may be right to use a low rate. Novo’s cash flows are large, its debt is modest and its balance sheet is rated AA. I do not think the risk is low enough, because 74% of sales sit in one molecule that is being repriced.

How good the business is and how good the price is are separate questions. The business is very good and has been for a decade. The stock has already fallen 73% from its June 2024 high, and a big fall is not a margin of safety. What matters is what is left to pay for. At $40.00 the market still pays for a business that earns about what it earns today, for ten years, and for more than my base gives it.

How the odds shrink

The bull needs many things at once. These are my judgments of each.

That comes to about 7% E (estimate). The six are not independent, since a strong pill helps the others, so the true figure is higher. If I raise every line to 85%, it is still 38% C (calculated). And the bull, after all that, is worth 22% more than the price. The base needs fewer things, and it is 30% below.

The margin of safety

• Price$40.00Close, Q4 2026 [1]
• Base value$28My estimate E (estimate)
• Margin of safety below$20the base value less 30% C (calculated)

I set it at the base value less 30%, the same rule as on the Lilly page, because one molecule is 74% of revenue and the value after 2036 rests on medicines not yet approved.²⁴ Below $20 my base returns about 13.5% a year and my bear 3.4%, so the owner is still paid if the bear arrives E (estimate). At today’s price there is no margin of safety in the base. The market may be right and I may be wrong, and the reason would be that Novo’s cash flows are safer, or its volume larger, than I can show today.

Where this could be in 5 and 10 years

This is the same model as above, read at later dates. The value at the end of 2031 or 2036 is the present value at that date, at 9.0%, of the owner cash that comes after it, less DKK 124B of net debt, on today’s 4,420M shares. Owner cash paid is the sum of each year’s cash to 2036, before interest on the debt. The return counts that cash as received each year and the 2036 value at the end E (estimate). It runs about half a point above the 4.7% in the reverse DCF, because here the debt stays flat instead of being charged from today C (calculated).

For the base to hold at five years, the 2027 list-price cut has to cost about what I assume. Sales are near DKK 350B in 2031, the margin is 37.5%, capital spending is down to DKK 28B, and owner cash reaches DKK 84B. It slips to DKK 76B by 2036 because the 2032 patent end takes the margin to 36%. That is why the value rises only from $28 to $32 to $36 while $25 is paid out. Even in 2036 the base is below the price today E (estimate).

The cases differ on three numbers. The bear has sales falling to DKK 218B and a 28% margin. The bull has sales at DKK 448B and a 42% margin. All three pay 5% of sales for deals and use the same 9.0% rate. At 7.3% the base equals the price, so the rate matters as much as the operating numbers. The likeliest way my base is wrong is net price after the 2027 cut. A 20% swing in sales every year moves today’s base from $28 to $22 or $35 E (estimate).

These are estimates of value, not forecasts of the share price. The market price can sit above or below value for years.

VIIThe case for the price

What most people miss

Capital spending is a peak, not a rate. Novo spent DKK 60.1B on plant in 2025 and plans about DKK 55B in 2026, against DKK 5.8B in 2020, and says it expects the spending to decline.¹²⁶ Each DKK 10B that does not recur adds about $0.34 per ADR of yearly cash C (calculated). The market is pricing cash flow at the peak of the build.

International Operations is half the company and growing. Its obesity care sales rose 40% in the first half, 57% in Europe and Canada.⁶ Much of the debate about Novo is about the United States. The other half is growing at the pace the US had two years ago.

The bull case

The company’s version comes from Mike Doustdar. On 4 August he said the Wegovy pill, early sales abroad and new launches had led Novo to raise its 2026 guidance.⁶ At the 21 September Capital Markets Day, STAT reported, he said Novo would launch at least five multi-blockbuster medicines by 2030 and do more business development.¹⁹ These are company claims.

The market-size argument comes from the rival. On its call, Lilly’s CFO said the company is reaching about 1 in 10 patients in the US, and the transcript gives 2 in 100 for the next figure, which reads as the rest of the world.²⁴¹⁷ My bull case needs that growth to continue and Novo to keep a large share of it.

I searched for a named investor or analyst with a verifiable 2026 bull case on Novo. I found no statement from Gerstner, Baker, Ackman, Wood, Gurley, Chamath or Sacks. The case below is mine, built from the company’s filings.

Against reality

If Novo hits the path the price needs at my 9.0% rate, with obesity care sales of DKK 116B in 2027 and DKK 189B in 2030, an operating margin near 40% and capex falling to DKK 30B by 2030, the price is possible E (estimate). That path sits 61% of the way from my base to my bull. Today’s numbers are some way off. Obesity care ran at DKK 88B a year in the first half, so 2030 needs 2.1 times that, and total revenue of DKK 372B is 23% above the adjusted twelve months C (calculated).⁶

The margin is the harder test. It was 44.6% in the half and 40.2% over twelve months, and the path needs about 40% through a US list-price cut of 35% to 50%.⁶

What a 10x would take. A tenfold price is $400 an ADR, a market value of $1.77T, or DKK 11.8T C (calculated). At 20 times earnings, against 11.7 times today, that needs $88B of net profit, or DKK 591B, 5.9 times the DKK 101B of the last twelve months C (calculated). At a net margin of 30%, a little below today’s 33%, it needs DKK 1.97T of revenue, 6.5 times today’s. Over ten years that is 20.6% a year for revenue, and 25.9% a year for the share price C (calculated). In the last ten years, 2016 to 2025, Novo’s revenue grew 12.0% a year C (calculated).¹²⁵ The ADR rose 8.3 times from $17.77 in October 2016, a market value of about $91B, to $146.91 in June 2024, and then lost 73%.¹¹⁰ A tenfold move from $177B would make Novo worth 39% more than Eli Lilly’s $1.27T on the Lilly page.²⁴ The Lilly page records a 19-fold rise for Lilly in ten years, from a far smaller base, and I know of no company that has gone up tenfold from this size on revenue growth like Novo’s.²⁴ It is not plausible, and I would not plan on it. The honest range of outcomes is the $12 to $49 above.

Where I come down

The capex peak, the cost cuts, International growth and the pill moved me. The 2027 list-price cut and the 2032 patent did not, and they set my 36% margin, so my value stays at $28. I would raise it if net price falls by less than 10% a year after the cut, if capex reaches DKK 45B in 2027, or if CagriSema and zenagamtide win share against retatrutide. A margin 2 points higher every year would make the base $30 E (estimate). I do not know what net price will do.

VIIIRisks, and what would change my view

1. Price. The US list price falls to $675 on 1 January 2027, Medicare pays under a government agreement, and the pill sells for $149 to $299 in cash.⁶ US adjusted sales fell 4% in the half.⁶
2. One molecule. Semaglutide was 74% of 2025 sales, and its US patent ends in 2032.⁴ Generics launched in India in the first quarter and the first have been approved in Canada.⁷
3. A stronger rival. Lilly guides about 40% growth and reports retatrutide data well ahead of Ozempic 2 mg in type 2 diabetes.¹⁵¹⁸ CagriSema did not clear non-inferiority to tirzepatide on blood sugar.⁶
4. Manufacturing. The FDA review of denecimig is extended because of remediation at a manufacturing facility, and the company gave no new date on 2 October.¹⁴
5. Capacity. Novo has built for growth that is now slower. It took DKK 3B of costs in the second quarter to shrink contracts.⁶
6. Litigation. Putative class actions over an IT incident disclosed on 11 June 2026, an antitrust suit by a compounding pharmacy, and generic filings against Ozempic in the US and formulation suits in India.⁶⁷
7. Research. Impairments of DKK 9.5B in 2024 and DKK 6.3B in the second quarter of 2026, and the ZEUS trial missed its goal.¹²⁶
8. Control. One foundation holds about 77% of the votes, and the board was replaced at an extraordinary meeting in November 2025.⁵⁹
9. Currency. Costs are in kroner and sales mostly in dollars. A 5% move in the dollar is worth DKK 5.0B of operating profit over twelve months.⁶

What would raise my base value: net price that falls less than I assume, capex that falls faster, a generic date later than 2032. What would lower it: a second year of US decline, a rejected CagriSema, or Lilly’s oral share rising above half. A lower price would not change my base. It would change the answer to the third way of getting hurt.