Eli Lilly

IThe business

Lilly does one thing: it discovers, makes and sells medicines. It reports one segment, sells in about 90 countries and employs about 50,000 people, 12,000 of them in research.² It is 150 years old. For most of that time it was an ordinary large drug company. Revenue was $18.3B in 2016 after removing Elanco, the animal-health business it spun off, and $34.1B in 2023. It was $65.2B in 2025, and the company now guides $91B to $92B for 2026.¹⁴¹²²⁴

Revenue in the twelve months to 30 September 2026 was $86.5B, up 46% on the same period a year earlier C (calculated). Net income was $27.6B after $8.1B of acquired research charges C (calculated). TTM on this page means those four quarters.²³¹¹ Almost all of the change comes from one molecule. Tirzepatide is sold as Mounjaro for diabetes and Zepbound for obesity, and it brought in $55.4B, 64% of the total C (calculated).²³

What it sells

• Incretins. These copy gut hormones that control blood sugar and appetite. Tirzepatide acts on two of them, GIP and GLP-1. Foundayo, or orforglipron, is a daily GLP-1 pill approved for obesity on 1 April 2026, with self-pay from $149 a month at the lowest dose. It brought in $184M in the quarter.²¹⁹⁴
• Older diabetes medicines. Trulicity, down 14% in the quarter, and Jardiance, which is a share of Boehringer Ingelheim’s sales. The insulins sit in this group too.³
• Oncology. Verzenio for breast cancer was $5.9B in the TTM C (calculated). Jaypirca, Inluriyo and Retevmo are the newer products.²³
• Immunology and neuroscience. Taltz, Ebglyss, Omvoh, and Kisunla for early Alzheimer’s disease, which grew 175% in the quarter from a small base.⁴

Volume up, price down

The business is selling far more of a medicine for a lot less per patient. In the third quarter volume rose 53% and realized price fell 15%. In the US volume rose 41% and price fell 7%, or 10% without a favourable change in rebate estimates, as management told it on the call. Outside the US volume rose 76% and price fell 29%, mostly because Mounjaro joined China’s national reimbursement list.³⁵

Mounjaro now sells more outside the US than inside it: $5.7B against $5.0B in the quarter, up 93% abroad.⁴ Selling in cash-paying markets is new for this industry. Management says self-paying patients stay on Zepbound nearly as long as covered ones.⁵

Who pays

Three wholesalers, McKesson, Cencora and Cardinal Health, were each 16% to 24% of 2025 revenue, but they are pipes. The real payers are insurers, employers, governments and now patients themselves.² The liability for rebates and discounts was $24.2B at September, up from $17.4B nine months earlier.³ That is about one full quarter of revenue, held on the balance sheet as money owed back to payers C (calculated).

The newest payer is Medicare. Under a voluntary agreement with the US government, Medicare patients got discounted Lilly obesity medicines from 1 July 2026 through 31 December 2027, and Lilly said eligible patients might get Foundayo for $50 a month.³¹⁹ Management says more than half a million people filled a prescription in the first quarter of the program, 7 in 10 new patients started on a Lilly medicine, and about 80% had not been treated before.⁵ The program opened to about 20 million seniors, so that is about 1 in 40 so far C (calculated).⁵ The same agreement lowers Medicaid prices and commits Lilly to more balanced launch prices across rich countries.³

Where

The US was $41.7B of nine-month revenue, 62%. Europe was $12.2B, China $2.6B, Japan $1.8B, and the rest of the world $8.9B, up 129% C (calculated).³

What comes next

Retatrutide acts on three hormones. In the TRIUMPH-1 trial the highest dose gave 28.3% weight loss at 80 weeks, against 2.2% on placebo, on Lilly’s efficacy measure, which treats patients as if they stayed on the drug. Counting everyone as assigned, it was 25.0% against 3.9%. 11.3% stopped for side effects, against 4.9%.¹⁸ Lilly plans a US submission in the first quarter of 2027.⁶ Eloralintide, an amylin drug, is in Phase 3 alone and in combination, and the tirzepatide combination gave 23.3% weight loss at 48 weeks in people with type 2 diabetes.³⁵ Management says three quarters of the clinical pipeline is outside obesity.⁵

Lilly spent $18.2B on deals in nine months to fill that pipeline: Centessa for sleep disorders, Kelonia for cancer cell therapy, AtaiBeckley for depression, Orna, Ventyx, Curevo and Ajax. It closed Merida in October for up to $2.9B.³

IIThe moat

Three things protect Lilly. It can make this medicine at a scale few can. It has the trial data. It has the patents. All three are strong today. The first and second last for as long as Lilly keeps spending. The third has a date on it.

Making tirzepatide is hard. It is a long peptide, filled into injector pens. Property and equipment rose from $17.1B at the end of 2024 to $32.2B at September, and inventory from $7.6B to $18.2B.²³ Since 2020 Lilly says it has committed more than $55B to manufacturing, and it broke ground in Houston for $6.5B and in Richmond for $5B this quarter.⁷⁵ Foundayo is a small-molecule pill and needs no pens, which makes the next wave cheaper to serve and easier for others to copy.

Evidence it is widening

Share. Lilly gained nearly 4 points of incretin sales outside the US in a single quarter, to about 58%. In the US about 6 in 10 obesity prescriptions and more than 7 in 10 injectable ones are Lilly’s. These are management’s figures from IQVIA data.⁵

Cost. Prices fell 14% in nine months and gross margin rose 0.7 points to 84.0%. Lilly credits a lower cost of production and product mix.³ A company whose margin rises while its price falls has room left.

Staying power. Management says patients stay on Zepbound about as long as on Trulicity, which it puts at 16 to 18 months.⁵ That is a company claim, and it runs against the common view that people quit these drugs within a year.

Evidence it is narrowing

Government. The US now sets the terms on a large part of this market. The Inflation Reduction Act lets Medicare set prices nine years after approval for medicines approved as drugs, and thirteen for biologics. Trulicity and Verzenio take government prices in 2028.³ Tirzepatide was approved in 2022, so on that clock its Medicare price is set around 2031 or 2032 E (estimate).

Generics. In July 2026 multiple generic companies filed to sell Mounjaro or Zepbound before some or all of the listed patents expire. Lilly sued in Delaware in August.³ Generic launches by other makers in India, Brazil and Canada did not cost it share, management says.⁵

Copies. The 10-Q still names mass compounding as a live problem.³

The protection is strong and it is dated. An owner at this price is paying for what comes after 2036 as well as what comes before it.

IIIThe competitive advantage

Lilly’s edge is simple. It has the better drug in a very large class of medicine, and more of it to sell. Novo Nordisk is the only rival with comparable scale today.

The pill row deserves the most attention. Novo’s pill passed 265,000 prescriptions in one week in July. Foundayo had over half a million in the whole of Q3. Four in ten new patients has not yet become four in ten of the volume.¹⁷⁵ Novo had the first obesity pill to market and it is ahead. Lilly says Foundayo’s new patients doubled in the quarter, and that about three quarters of Medicare patients chose an injectable.¹⁷⁵ On the Q2 call analysts called the US launch slower than expected.⁶ I do not read one quarter as a verdict. A pill is where this market could start to look like ordinary primary care, with cheap competitors and buyers who see one product as much like another.

Where Lilly wins

It wins where efficacy decides: patients who want the most weight loss, doctors treating diabetes, and any buyer choosing between injections. It wins where supply decides, because its factories are built. And it wins in cash-paying markets, where a brand and a price list matter more than a formulary.⁵

Where it loses

It can lose in the pill if price and convenience matter more than the last few points of weight loss. It can lose with payers who treat the class as interchangeable. The 10-K says self-insured US employers must generally opt in to obesity cover, and some are opting out.²⁵ The 10-K also names a newer threat: Chinese developers now license medicines to multinationals and compete directly.²

The advantage is real and measurable today. What I cannot judge is what a daily pill does to the price of the whole class by 2030.

IVManagement

David Ricks has run Lilly since 2017 and chairs the board. He was 58 at the last 10-K and rose through the company. Under him revenue went from about $20B in 2017 to a guide above $91B, and most of that came from incretins. I give him credit for the bet and for building factories ahead of orders.²

Ricks owns 815,535 shares, worth about $1.16B at the close, or 412 times his salary on the proxy’s measure, and the proxy says none of the listed holdings is pledged. Company policy bars hedging or pledging.⁸¹ His own money is in, and I found no Form 4 showing a sale by him since the proxy date.³⁵ His 2025 pay was $36.7M, 293 times the median employee.⁸

Stock awards reward the share price over three years, and the 2036 patent falls outside every measure. The bonus rewards launches, which fits a period like this one. It also sets aside research charges from deals done during the year, so buying pipeline costs the bonus less than building it would. I see no measure tied to return on the $18.2B spent on deals this year. That does not make the deals bad. It means nothing in his pay asks him to prove they were good.⁸³

The price is already about 40% above the 2025 award’s target C (calculated). The payout is capped at 200%, so once the stock is far enough above target, as it may already be, that award stops paying for more.⁸¹ A cap can cut either way. It removes the reward for a short-term push as well.

Who can hold them to account

The board is staggered and supermajority rules still stand. In May, 665M votes favoured ending the staggered board, against the 756M needed, 80% of shares outstanding C (calculated). The supermajority proposal failed the same way.⁹ Owners holding a majority wanted the change, and it fell short of the rule. Pay passed with 96% of votes for and against C (calculated).⁹ An independent chair proposal got about a third.⁹

Capital over a full cycle

From 2016 to 2025 Lilly earned $42.8B of owner free cash flow C (calculated). It spent $32.5B on dividends, $20.1B on buybacks and $28.4B on deals.^1412216 Debt rose from $10.3B to $42.5B, and to $49.5B at September.¹⁶³ So the cash paid out and spent on deals, $81.0B, was almost twice what the business produced C (calculated). For most of that decade Lilly was paying for its future with borrowed money. That bet paid off.

The dividend per share rose from $2.05 declared in 2016 to $6.23 in 2025, 13.1% a year C (calculated), and it is now $1.73 a quarter.¹⁶³ At this price it yields 0.5% C (calculated).

Ricks has real money in, a clear record on the bet that mattered, and pay that looks three years out at a company whose main question is ten years out.

VFinancials

Earnings are excellent. Cash is lower than earnings, and the reasons are worth an owner’s time: heavy building, large deals and more tax paid than booked.

Gross margin was 83.6% in the TTM, from 77.3% in 2016 C (calculated). Research ran at 17.8% of revenue and selling and administration at 14.6%, so performance margin, which is gross margin less those two, was 51.3% C (calculated). It was 17.9% in 2016 C (calculated).¹⁴³² The company guides 52.0% to 52.5% for 2026 and reported 54.2% on a non-GAAP basis in Q3.⁴⁵ Each added dollar of revenue in the TTM brought about 66 cents of performance income C (calculated).²³¹¹ My base has about 59 cents from 2026 to 2030 E (estimate), because I expect price to keep falling.

What an owner can really take out

Reported net income is $27.6B. That is the wrong number for an owner. It is after $8.1B of acquired research, which is lumpy, and before a lot of cash that goes out elsewhere. So I walk it two ways.

So an owner takes out about $20.1B, 27% less than reported earnings E (estimate). The stand-still line is my estimate, and I cannot know it well. Depreciation understates the true cost of keeping new plants running, and a drug company that stopped buying pipeline would be liquidating itself slowly. Deals at 5% to 9% of revenue give $18.4B to $21.8B, and each extra $1B a year of stand-still building takes $1B off.²³ I would rather show that range than a false decimal. Stock pay is small, 0.8% of revenue, far below the technology pages on this site C (calculated).

Management says about $2B of 2026 revenue will not recur, including about $900M in Q3, mostly Jardiance milestones and rebate estimate changes. No Jardiance milestones remain.⁵⁴³

Tax

Lilly paid $10.8B of income tax in cash in 2025 against $5.1B of expense. The 10-K puts the rise mostly on Ireland, where it makes more product.² Prepaid taxes reached $15.1B at September.³ That is cash already out of the door ahead of the income statement. The IRS opened 2022 to 2024 in October 2026, on top of 2019 to 2021, and 2016 to 2018 pricing is in international talks.³ My cases use 20%, the top of the 19% to 20% guide.⁴

Balance sheet

Debt was $49.5B and cash $4.9B at September, so net debt was $44.6B, 1.0 times the TTM’s performance income C (calculated).³ That is comfortable. It is also a change: Lilly added $7.0B of debt in nine months while earning record profits, because deals and factories took more than the business produced. Interest expense was $1.06B in nine months, up from $672M. It has $10.1B of unused credit lines. Contingent payments for deals carry a fair value of $2.9B.³ Goodwill and intangibles rose to $28.0B from $12.4B C (calculated).³

Receivables were $22.1B and inventory $18.2B, up 24% and 32% in nine months against revenue up 39% year on year C (calculated).³ Finished goods are only $3.0B of inventory. Work in process, the part that takes longest to make, is $10.7B.³ Working capital took 21% of each added dollar of TTM revenue, and I charge 20% C (calculated).²³¹¹

VIValuation

What would ruin it

An owner can get hurt here in three ways, and only two of them are about the company.

1. The price of the class collapses. Governments, employers and cheap pills push net price down faster than patients arrive. Lilly already took 15% off in a quarter. Medicare cover ends on 31 December 2027 unless something replaces it, and the US wants prices aligned with other rich countries.³ The drug stays good and the business earns far less on it.
2. A safety or patent shock to one molecule. 64% of revenue is tirzepatide C (calculated). Lawsuits are consolidated in two federal multidistrict cases in Pennsylvania, over gastrointestinal injury and a form of vision loss, and generic companies are challenging the patents.³ One bad long-term finding would hit most of the earnings at once.
3. The owner pays too much. This needs nothing to go wrong at Lilly. The business can do everything I expect and the investor can still earn little, because the price already assumes it. At today’s price, this is the most likely of the three on my numbers.

The first two sit in my bear case. The third is what the rest of this section measures.

What the price implies

At $1,417.54 Lilly is worth $1.27T on 893.4M diluted shares C (calculated). That share count leaves out the 50M shares in the employee benefit trust, which the 10-K does not treat as outstanding, and it is why the cover’s 940.7M is not the right number.¹³² The stock rose 10% in two days on the results and 32% this year, and the lowest close of the past year was $851.21 in April.¹

Start with earnings. The price is 30.0 times Nasdaq’s 2027 consensus of $47.23 and 25.4 times the 2028 mean of $55.72, a mean of only three estimates C (calculated).²³ Every peer is far lower.

A premium is earned. Lilly is growing 39% and these companies are not. The harder question is how long the growth runs and what is left when the patent ends.²⁴

The reverse DCF starts from the $20.1B of owner cash. I solve for one growth rate for ten years, take a 15% cut for the patent, then grow 3%, at my 9.0% rate. The price needs 22.1% a year, to $149B in year ten C (calculated). Owner free cash flow grew 15.6% a year from 2019 to 2025 C (calculated), and my base grows 12.7% a year and my bull 19.0% from the same start E (estimate).¹⁴²³ At a 7% rate the price still needs 16.0% E (estimate).

The price asks owner cash to grow faster for ten years than my bull case does, and faster than the last six years did.

The discount rate

I use 9.0%, the hurdle on my other large, established companies. The capital asset pricing model gives far less. With a 4.94% ten-year Treasury and Damodaran’s 4.20% premium, Lilly’s month-end betas against the S&P 500 of 0.37 over ten years, 0.43 over five and 0.51 over three give 6.5% to 7.1% C (calculated).^2021221 Those betas describe a stock that happened not to move with the market while one drug took it up 19 times. They say little about a company with 64% of revenue in one molecule. I also note that Lilly’s own board uses a 9.48% cost of capital in the executive stock award.⁸ That is the most useful single number I found. If management is paid to beat 9.48%, 9.0% is not harsh. On it the base is $683 E (estimate).

The price is above every cell in that table. The base equals the price at a 6.3% rate E (estimate). This is the plainest way to say what the market believes. It treats Lilly’s cash flows as nearly as safe as a long government bond plus 1.3 points. It may be right about the cash flows. I am not willing to discount a single-molecule business that way.

What I cannot judge

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

• The net price in 2032. It depends on Medicare rules not yet written, on what the agreement with the government means after 2027, and on pills. The 10-Q says itself that the outcome is uncertain.³ A 20% swing in incretin revenue moves the base from $619 to $869 E (estimate).
• What retatrutide and the amylin drugs are worth. They have good data and no approval. The US filing is planned under the biologic pathway, and Ricks described a first court decision on that as mixed.⁶
• Long-term safety in tens of millions of people.
• What the $18.2B of deals will return. Most of it is early-stage, and I have no way to value a shingles vaccine, an in-vivo cell therapy and a psychedelic against each other.³

Three cases

I build each case from incretin revenue, other revenue, performance margin, tax, capital spending, 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, because a company losing a patent every decade must buy or build the next one. Lilly’s deals were 9.2% of revenue over ten years and 14.7% from 2023 to September 2026 C (calculated).¹⁴¹²²³ I use 7% in the base, and research stays inside the margin as well.

The year 2036 needs care. The US compound patent ends that year, Europe’s in 2037 and Japan’s in 2040.² A ten-year model ends right at the cliff, and 3% growth forever on a peak would be wrong. In the base I assume tirzepatide is about $84B of 2036 revenue, 40% of the total E (estimate), that it would lose about two thirds of that, and that successors win back most of it. The net cut to year-ten cash flow is 15% E (estimate). The deals I charge pay for those successors. The cut is 30% in the bear and 5% in the bull. It is a judgment, and each 15 points of it is worth about $80 a share E (estimate).

The base is a good outcome. Incretin revenue grows 2.6 times from 2026 C (calculated). If net price falls 6% a year, a pace I chose as a middle reading, the number of patient-years on Lilly medicines has to rise about 4.8 times E (estimate). With 1 in 10 reached in the US and 2 in 100 elsewhere, management’s figures, that is plausible.⁵ Base earnings are about $52 in 2027 and $60 in 2028 before interest and deals E (estimate), above the $47.23 and $55.72 consensus, which is after them.²³

The bear has the price of the class fall faster than volume grows, Medicare cover shrink after 2027, a pill market that looks like generics, and deals that cost more and yield less. Revenue still rises to $120B. Lilly stays a healthy company, and an owner at $1,417.54 still loses 9.7% a year E (estimate).

The bull has retatrutide approved early, Medicare cover made permanent, cash-pay markets abroad working, and the new pipeline delivering $78B of other revenue. Revenue reaches $320B, nearly five times the $65.2B of 2025 C (calculated). It is still below the price, by 6%.

Weighted 25%, 50% and 25%, the cases give $772 a share E (estimate). At 20%, 50% and 30% the figure is $826, and at 30%, 50% and 20% it is $719 E (estimate). Against the price the skew is −81% to my bear and −6% to my bull E (estimate). A mid-year convention adds about $35 to the base E (estimate).

A ten-year Treasury pays 4.94%, and my base return on Lilly at this price is 1.3%, with my bull at 8.3% E (estimate).²⁰ The fairest comparison on this site is Mastercard, another excellent business at a high price, with little capital need and no patent cliff: my base there returns 8.4% a year. NVIDIA returns 9.7% on the same measure and Berkshire about 7.0%.³¹³⁰³² Lilly’s owner cash yields 1.6% today C (calculated), 3.3 points below the Treasury, so growth has to make that up first.

What closes the gap

Operating changes do not close it. Incretin revenue would need to be 2.1 times my base every year, $253B in 2030 and $326B in 2036 E (estimate). Only the discount rate closes it alone. The market may be right to use a low rate. These cash flows have surprised upward all year, with guidance up about $10B, and lower-risk cash deserves a lower rate.⁵ I do not think that risk is low enough.

How good the business is and how good the price is are separate questions. The business is among the best I have written up. This looks like a great business at a price that has already counted that and more.

How the odds shrink

The bull needs many things at once, and these are my judgments of each.

That comes to about 9% E (estimate). The six are not independent, since a strong drug helps most of them, 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 6% less than the price.

The margin of safety

• Price$1,417.54Close, Q4 2026 [1]
• Base value$744My estimate E (estimate)
• Margin of safety below$521the base value less 30% C (calculated)

I set it at the base value less 30%, more than the 25% I use for broader businesses, because one molecule is 64% of revenue and the value after 2036 rests on medicines not yet approved. Below $521 my base returns about 13.6% a year and my bear 0.9%, so the owner is paid even if the bear arrives E (estimate). That is 39% below the lowest close of the past year.¹ At today’s price there is no margin of safety in any part of my range. The market may be right and I may be wrong, and the reason would be that Lilly’s reach is far larger, or its cash far safer, than I can yet show.

Where this could be in 5 and 10 years

In the base, today’s $744 becomes about $922 in December 2031 and $1,031 in December 2036. For year 5 to hold, incretin revenue has to reach $132B in 2031 and the performance margin has to stay at 55%, so owner cash reaches $53B a year. For year 10 to hold, Lilly has to replace the incretins as the 2036 patent ends. I take owner cash down 15% at that point and grow it 3% after. That terminal value is 58% of today’s base and nearly all of the year-10 value.E (estimate)

The bear and the bull differ on growth and margin. In the bear, incretin revenue peaks at $94B in 2031 and falls to $84B by 2036, and the margin drops to 42%, so the value stays near $290 and owners receive $262 a share in ten years. In the bull, incretin revenue reaches $242B in 2036 and the margin ends at 56.5%, which gives $2,047. Capital needed is $12.5B to $14.5B a year to 2031, then 7% of revenue, in all three. The discount rate moves them all: at 8% my base is $906 today, not $744; at 10% it is $627.E (estimate)

The most likely way the base is wrong is incretin revenue. If it comes in 20% lower every year, today’s value is $619, not $744.C (calculated)

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

VIIThe case for the price

What most people miss

Lilly’s prices fell 14% in nine months and its gross margin rose 0.7 points.³ Most drug companies cannot do that. It means unit cost is falling faster than price, at least for now.

Retatrutide is planned for filing as a biologic.⁶ That matters. Under the rules in Lilly’s own filings, a biologic gets twelve years of US data protection and thirteen years before Medicare sets a price, against five and nine for a drug.²³ If Lilly wins that argument, its next obesity medicine could keep its price for four years longer than tirzepatide.

People stay on these drugs longer than the common story says. Management puts Zepbound persistence close to Trulicity’s 16 to 18 months, even for self-pay patients.⁵

The bull case

Lucas Montarce gave it on the call. The US has reached about 1 in 10 patients and the rest of the world about 2 in 100, so he sees years of runway, and he read this year’s raises as demand and not patients pulled forward.⁵ Medicare added more than half a million patients in a quarter, most of them new to treatment.⁵ David Ricks argues cover will widen because a University of Southern California study estimates it would save Medicare money.⁵ These are company claims.

Investing.com reports that BofA raised its target to $1,497 on 25 times higher 2028 earnings, cites results that stayed strong after one-offs, and says Lilly dominates the obesity market.²⁵ It also reports Bernstein kept a $1,450 target and pointed to Medicare, Foundayo abroad and diabetes, while flagging about $2B of 2026 items that will not recur.²⁶ ARK’s 13F listed 77,254 shares at 30 June, 0.6% of its portfolio, from 12,676 at 31 March. Duquesne’s listed 12,380, from none. These are positions only, and I found no stated view from either.²⁷²⁸

Against reality

If Lilly hits my bull path, with incretin revenue of $110B in 2028 and $242B in 2036, and the market accepts about 8.7% as the right rate, the price is possible E (estimate). Today’s numbers are a long way off. Incretin revenue ran at $65B a year in Q3 C (calculated), so 2036 needs 3.7 times that C (calculated), and at lower prices. Margin needs to reach 57.5%, against a 2026 guide of 52.0% to 52.5%.⁴ Capital spending has to peak near $14.5B, and it was $11.0B in the TTM C (calculated).

BofA’s $1,497 implies about $60 of 2028 earnings C (calculated). My base has about the same E (estimate). We do not disagree on 2028. We disagree on whether 2028 earnings deserve 25 times with 2036 eight years out. At 25 times, Lilly would trade far above the peers in the table above.

What a 10x would take. A tenfold price is $14,175, a value of $12.7T C (calculated). At 25 times earnings that needs $507B of net income. At a 40% net margin, well above today’s 32%, it needs $1.27T of revenue C (calculated). By 2036 that is 26% a year for ten years in the share price C (calculated). Revenue would be 15 times today’s $86.5B, and four times my bull. NVIDIA’s TTM revenue is $303.0B.³⁰ NVIDIA is the best recent case of a very large company compounding, and from about $1T in June 2023 its price has risen about 5 times C (calculated).²⁹ Lilly itself did 19 times in ten years, from a far smaller base C (calculated).¹ I found no company that has gone up ten times from above $1T. For Lilly it is close to impossible.

Where I come down

The Medicare numbers and the cost figures moved me: I give the base more volume than I would have a year ago. They did not move the price question, because $744 already has 2028 earnings at the bulls’ level. I would raise my values if retatrutide is accepted as a biologic and a lasting Medicare decision arrives; incretin revenue 20% above my base every year would make the base $869 E (estimate).

VIIIRisks, and what would change my view

1. Price. The agreement with the US, most-favored pricing in any form, and pills. Realized price fell 15% in the quarter.³
2. One molecule. Mounjaro, Zepbound and Foundayo were 66% of nine-month revenue, and the 10-Q says incretin volume swings can have a disproportionate effect on results.³
3. Medicare. The program ends 31 December 2027, and uptake after that is unknown.³
4. Patents and generics. Tirzepatide challenges filed in July 2026; trial on Verzenio in May 2028.³
5. Patent roll-off before 2036. Trulicity in 2027, Verzenio in 2031, Taltz in 2030, the end of the main Jardiance collaboration on 31 December 2028, and government prices for Trulicity and Verzenio in 2028.²³
6. Safety and litigation. Two federal multidistrict cases on incretin injuries, plus insulin pricing and 340B suits. Lilly is mostly self-insured for product liability.³
7. Deals. $18.2B in nine months, mostly early-stage, and a partial impairment of one development asset has already appeared.³
8. Capacity. Lilly is building for demand it expects years from now. If the price falls first, those plants earn less.³
9. Tax. The IRS has opened 2022 to 2024 and prepaid taxes stand at $15.1B.³

What would raise my base value: a biologic ruling for retatrutide, lasting Medicare cover and a lower price decline. What would lower it: a class-wide safety finding, a successful patent challenge, or another year of deals near $18B. A lower price would not change my base, but it would change the answer to the third way of getting hurt.