One company in our library won the drug war of the century — four head-to-head trial victories, the first trillion-dollar healthcare company in history — and cut its price 13% in the same quarter.
That sentence is healthcare, complete. The science was won decisively. The economics of winning were set by somebody else.
Across our eight healthcare reports the average score is 6.48, our second-lowest sector, with a spread from 4.8 to 7.8. That is not because these are bad companies. Several are magnificent. It is because the sector has a structural feature that almost no other sector has.
The customer does not set the price, and usually does not pay it
In every ordinary business there are two parties: someone who wants a thing, and someone who sells it. The price emerges between them.
In healthcare there are at least four. The patient consumes. A doctor prescribes. An insurer or a government decides what will be paid. A regulator decides what may be sold at all. The company that spent a decade and billions inventing the product negotiates with three parties, none of whom is the person receiving it.
This is why the sector produces the specific pattern our scores keep finding: outstanding gross margins, outstanding returns on capital, and a persistent inability to convert either into a durable, predictable stream of shareholder value.
Winning the science is necessary, and nowhere near sufficient. Three parties who are not your customer decide what winning is worth.
Healthcare is not one sector, it is three
They fail in completely different ways
| What it sells | How it fails | Our range | |
|---|---|---|---|
| Discovery | Molecules, protected by patents with dates on them | The patent expires and nothing replaced it | 4.8 → 7.8 |
| Tools & services | Equipment and services to whoever is doing the research | Overpaying for growth — acquisitions that never earn their cost | 5.6 → 6.2 |
| Payers & distribution | Scale applied to the plumbing, under heavy regulation | A regulator, a lawsuit, or a political cycle | 6.0 |
Look at the failure column, because it is the practical part. These three do not just earn differently — they break differently, and no single analytical checklist covers all three. A discovery business is judged on its pipeline and its expiry dates; a tools business on the returns it earns on acquisitions; a payer on its relationship with regulators and the courts.
What each of ours actually shows
- Eli Lilly (7.8) — won four head-to-head trials, became the first trillion-dollar healthcare company, tripled revenue in a decade. And cut prices 13% in the quarter it won. The prize arrived smaller than the victory.
- Novo Nordisk (7.3) — down about 63%, with 82% gross margins and a 66% return on equity intact. The economics never broke. The competitive position did, and the market priced the second thing.
- UnitedHealth (6.0) — recovered ~80% off the lows with earnings re-accelerating, against a valuation model saying overvalued, a federal criminal probe, and Berkshire already gone. Three different signals, all legitimate, pointing opposite ways.
- Thermo Fisher (5.6) — $30 billion of acquisitions in six years, free cash flow that went from $6.82bn to $6.29bn, and GAAP earnings per share still below 2021. Growth was bought; it has not yet arrived.
A business with 82% gross margins and a 66% return on equity is, on the numbers alone, among the finest we have examined in any sector. It fell 63% anyway — not because the economics deteriorated, but because a rival's molecule worked better.
In most industries a competitor with a better product takes share slowly, over years, against switching costs and habit. In pharmaceuticals, a head-to-head trial result can move prescribing behaviour in months. The moat is the molecule, and molecules can be beaten in public, on a schedule.
How to actually approach it
Before any ratio. A tools company and a discovery company examined with the same checklist will both look wrong, in opposite directions.
Published, and ten minutes of work. Then ask what the last decade of research spending actually produced — that is the only available evidence about what the next decade will.
If the answer is a government, an insurer, or a formulary committee, then the gross margin you are admiring exists at someone else's discretion — and discretion gets exercised, usually at the least convenient moment.
A discovery winner, a discovery loser with intact economics, and a payer under investigation. Same sector, three different questions.
Each with the pipeline, the pricing and the regulatory position worked through.
Frequently asked
Why do healthcare companies score poorly in your research?
Our eight healthcare reports average 6.48, our second-lowest sector, and the spread runs from 4.8 to 7.8. The pattern behind it is consistent: the economics look superb — high gross margins, high returns on capital — and then three parties who are not the customer get a say in the price. Winning the science is necessary and nowhere near sufficient.
Is Eli Lilly or Novo Nordisk the better investment?
We scored Lilly 7.8 and Novo 7.3 at the time of our July 2026 reports. Lilly had won four head-to-head trials and become the first trillion-dollar healthcare company; Novo was down about 63% with 82% gross margins and a 66% return on equity intact. That is the sector's cruellest joke — the winner cut its price 13% in the quarter it won, and the loser still runs extraordinary economics.
What makes UnitedHealth different from a drug company?
It is not a discovery business at all — it is scale applied to the plumbing of American healthcare, which is why it earns differently and fails differently. As at our July 2026 report it had recovered about 80% off its lows with earnings re-accelerating, against a discounted cash flow that said overvalued, a federal criminal probe, and Berkshire already out of the position.
How should I approach healthcare as an investor?
Separate the three business models before comparing anything. Discovery businesses live and die by patents with dates on them. Tools and services businesses sell to the researchers and are indifferent to which drug wins. Payers and distributors are regulated scale operations. They share a sector label and essentially nothing else.