Amgen Inc · Moat analysis

Report previewAMGN.USReport date Last updated
← Back to reports
Report contents
01

Industry landscape

Where the company competes: separate franchise fights, each with two to five rivals

Amgen sells branded prescription biologics and small molecules, and it does so in roughly a dozen molecule-level markets that price, compete and lose exclusivity on separate clocks. That fact shapes everything else about the company's competitive position. There is no single "biotech market" where Amgen holds a share: Repatha fights for cholesterol patients against two other PCSK9 agents, Prolia now fights copies of itself, and Tepezza faces no approved rival in thyroid eye disease. Amgen reported $33.4 billion of revenue in 2024, up 19%, the first full year to include Horizon Therapeutics.

The value chain starts in Amgen's own discovery labs and clinical trials, runs through in-house biologic manufacturing, and reaches patients through wholesalers and specialty pharmacies. The decision that sets the economics sits further downstream. Physicians write the prescription, but commercial payers and pharmacy benefit managers in the United States, and national health systems elsewhere, decide the net price. Repatha shows the mechanism: unit volume rose 43% in 2024 while net price fell 10%, so growth of 36% to $2.2 billion came from selling more units at lower realized prices. A franchise can win patients and still lose revenue per patient.

Geographically, the arena is global to register in and American to earn in. Amgen sells across the US, Europe and Japan, but the US is its largest market, and price formation differs by region: US net prices emerge from rebate negotiations and formulary placement, while European and Japanese prices are administratively set. Manufacturing scale and sales-force breadth travel across borders; a contract won with a US payer does not.

FranchiseReported productsFY2024 revenue (USD bn)Share of totalPrincipal rivals
Bone healthProlia, Xgeva, Evenity8.1624%denosumab biosimilars, bisphosphonates
InflammationEnbrel, Otezla, Tezspire6.4219%Dupixent (Regeneron/Sanofi), Humira and Skyrizi (AbbVie), Nucala (GSK), Fasenra (AstraZeneca)
Oncology and hematologyKyprolis, Blincyto, Lumakras, Imdelltra3.1910%CAR-T (Novartis, Gilead, Bristol Myers), checkpoint inhibitors (Merck, Bristol Myers)
Rare diseaseTepezza, Krystexxa3.049%none approved in the labeled indications
CardiovascularRepatha2.227%Praluent (Regeneron/Sanofi), Leqvio (Novartis)

Those products account for about 69% of reported revenue. Uplizna and Tavneos, also acquired with Horizon, are not broken out in the figures above.

Market structure: concentrated per molecule, exposed where patents lapse

Concentration is high at the level that matters for pricing. PCSK9 has three credible sellers, severe asthma four, and refractory gout and thyroid eye disease one each. That is the shape of a market where a new entrant must spend years and several hundred million dollars on trials before it earns a single prescription. Amgen's own history illustrates the cost of admission: Imdelltra reached traditional FDA approval in November 2025 only after Phase 3 survival data, following an accelerated approval. Nothing about that path is quick, and the resulting concentration supports price in a way that, say, generic statins never could.

The barrier has a seam. Protection rests on patents and regulatory exclusivity, not on customer habit or switching costs, and when those lapse the same regulatory machinery that kept rivals out ushers copies in. Denosumab is the live test. At least eight denosumab biosimilars won approval or launched between 2024 and early 2026, and Amgen guided in 2025 to accelerating erosion of Prolia and Xgeva in the second half of that year. The franchise is still 24% of revenue, so the erosion is not a rounding error. Enbrel, down 10% in 2024, shows the same pattern further along.

Inflammation is the crowded exception. Enbrel, Otezla and Tezspire compete in a field where AbbVie, Regeneron and Sanofi, GSK and AstraZeneca all field approved products, and where Tezspire at $972 million is the small participant against Dupixent. That franchise looks more like a market with several strong incumbents than one with a protected leader.

The structure therefore cuts two ways. Diversification is real: no single product exceeds about 13% of revenue, so no one rival's win damages the whole company, and the fast-growing assets, Repatha, Evenity, Tezspire and Blincyto, all grew at double-digit rates in 2024 and offset declines elsewhere. The vulnerability is concentrated in the one place where the barrier has already failed, the denosumab franchise, and in any other molecule approaching its exclusivity date. Whether the rare-disease assets stay uncontested is the open question: their protection rests on clinical data and specialized manufacturing, and a single approved competitor in thyroid eye disease would change the picture. So far the evidence only shows that none has arrived.

02

Competitive advantage

Does the company have a moat? Yes, a narrow one with an expiry date

Overall rating: Moderate. Amgen has real protection in the markets where its patents still run, and four decades of evidence that it can discover, test and manufacture biologics that payers will pay for. What the record does not show is that those returns survive the loss of exclusivity. The two tests that settle the question split: concentration in Amgen's molecule-level markets is high and has stayed high, but returns on capital fall short of the level that marks a durable advantage once the price paid for acquisitions counts as invested capital.

That split is what separates a structural moat from good execution. If Amgen earned well above its cost of capital on every dollar deployed while holding share through patent cliffs, entry barriers would be doing the work. If fat margins drew in rivals and vanished within a few years, the edge would be management skill. Amgen sits between: rivals cannot enter while a patent runs, and they enter the quarter it lapses. That is a barrier with a scheduled end date.

Market position: first in several molecule markets, dominant in none

Market position: Fairly Strong. Amgen competes in about a dozen molecule markets that share a sales force but not a competitive dynamic. It holds the largest PCSK9 franchise, with Repatha at $2.2 billion of 2024 sales against smaller Praluent and Leqvio. It is the only approved seller of Tepezza for thyroid eye disease and of Krystexxa for refractory gout. Until 2025 it was the sole marketer of denosumab.

That word "sole" is the pattern worth noticing. Where Amgen stands alone, the exclusivity is legal and temporary. Where the field has several sellers, Amgen is one of them: Enbrel competes with AbbVie's Humira and Skyrizi, with Regeneron and Sanofi's Dupixent, and with a half-dozen other agents, and its revenue has declined for years.

Amgen does not disclose share by molecule, and no independent series covers most of these markets, so ranking has to rest on reported revenue and on the count of approved rivals. On that evidence, no single product exceeds roughly 13% of revenue. That helps resilience and weakens the market-power case. There is no franchise where a 40%-plus share and stable pricing reinforce each other.

Stability: share holds while patents hold

Stability of position: Moderate. Denosumab is the test case in progress. Prolia and Xgeva lost US patent protection in early 2025, several biosimilar versions have launched since, and the franchise, 24% of 2024 revenue, is now shrinking by roughly a third year over year.

The rule of thumb is that a five-year average share shift of more than 5 percentage points argues against real entry barriers, while under 2 points argues for them. A franchise of that size falling that fast clears the first threshold on its own. Enbrel, a decade past its own cliff, shows where that ends: steady, compounding decline.

The offset is that Amgen keeps refilling the portfolio. Repatha, Evenity, Tezspire and Blincyto all grew at double-digit rates in 2024, and total revenue rose from $26.0 billion in 2021 to $36.7 billion in 2025, about 9% a year. Company-level share looks stable, but it is stable because new products are pushed into the gap. That is a refresh engine, not a defended position.

Profitability: returns fall below the moat threshold once goodwill counts
Fiscal yearRevenue (USD bn)Gross margin (%)Operating margin (%)Net margin (%)Free cash flow (USD bn)ROIC incl. goodwill (%)ROIC excl. goodwill (%)
202125.9875.229.422.78.3814.220.4
202226.3275.736.424.98.7915.823.0
202328.1970.028.023.87.368.511.2
202433.4261.521.712.210.3910.013.7
202536.7470.829.121.08.1015.521.7

Core returns: Moderate. ROIC here is an estimate built from reported figures, with invested capital taken as total assets less current liabilities and NOPAT as EBIT taxed at each year's effective rate. That method treats short-term debt as operating, so it modestly understates returns.

Reported ROE is unusable: equity was $8.7 billion against $90.6 billion of assets at the end of 2025, a residue of buybacks, and figures of 68% to 136% describe the capital structure rather than the business.

After-tax ROIC averaged about 12.8% over 2021-2025 including goodwill and 18.0% excluding it. The gap is the price paid for Otezla and Horizon, which left $18.7 billion of goodwill on the balance sheet. Both numbers answer a fair question. The 18% shows the operating businesses earn attractive returns. The 12.8% shows what shareholders got for the capital committed. Either clears a cost of capital near 7%-8%, so some barrier exists, but only the ex-goodwill figure reaches the 15%-25% range that marks a durable one, and the five-year average leans on 2021-2022, before Horizon closed in October 2023. Purchase accounting also depressed 2023 and 2024, since amortization of acquired intangibles runs through operating income. That is an accounting artifact, though the cash still left.

Margins: 70% gross, and half of it handed back

Margins: Fairly Strong. Gross margin ran from 61.5% in 2024 to 75.7% in 2022 and was 70.8% in 2025. Operating margin ranged from 21.7% to 36.4% and net margin from 12.2% to 24.9%. This is standard branded-biopharma economics: the marginal cost of producing a biologic is small next to the fixed cost of discovering, testing and selling it. Amgen's gross margin does not stand apart from large peers, so the industry-wide level says more about approved molecules than about this company.

What is company-specific is where the gross margin goes. Sales deductions of rebates, chargebacks and returns absorbed about 53% of gross product sales in 2025, based on the company's own reconciliation, up from roughly the low 40s in 2019. On $36.7 billion of net product sales, that is $42.0 billion handed back. Volume grew 13% in 2025 while net price fell 3%, so the growth comes from selling more units at lower realized prices. That is the pricing-power test, and Amgen fails it.

The moat is moderate, and return on capital is the binding constraint

Market position is real but fragmented and expiring: Amgen is first in the markets where exclusivity is ending and one of several where it is not. Profitability clears the cost of capital but only reaches the moat band once goodwill is stripped out. A company that passed both tests would rate Fairly Strong or better. Amgen passes one and a half.

Where the advantage comes from and how long it can last: legal fences, not customer locks

Three sources carry what protection Amgen has: regulatory exclusivity, scale in development and manufacturing, and the inertia of chronic biologic prescribing. They differ sharply in durability, and the most visible one is also the one with a date certain on it.

Patent and regulatory exclusivity: Fairly Strong, and on a clock

Regulatory protection: Fairly Strong while it runs. The causal chain starts with the cost of admission. Getting a biologic approved means preclinical work, three phases of trials and a manufacturing plant that passes inspection. Imdelltra reached traditional FDA approval only in November 2025, after an accelerated approval, once Phase 3 survival data arrived. That path costs hundreds of millions of dollars and most of a decade, which is why molecule markets end up with one to four sellers.

While exclusivity runs, no substitutable copy exists. Payers can negotiate rebates, but they cannot play one identical molecule against another, and that is the direct source of a 70% gross margin. Cash from protected molecules then funds the next round of trials and business development, which is how the model compounds.

The fence guards against copies, not against competition. A rival with a different mechanism can take the market: an oral drug, a better antibody, a combination. Inflammation has drifted toward oral and next-generation agents, and the cholesterol field has alternatives with different modes of action. Even inside the exclusivity window, price erodes, which is why Amgen sold 13% more volume for 3% less net price in 2025. And expiry is always known in advance. Denosumab's US patents ran out in February 2025 and a quarter of revenue began falling by a third within months. Whatever the moat is worth, its half-life tracks the remaining patent term.

Scale in development and manufacturing: Moderate

Scale: Moderate. In 2025, $36.7 billion of revenue produced $26.0 billion of gross profit, leaving about $15.3 billion of operating expense across R&D, selling and administration. The fixed assets behind that spending are the trial organization, the plants with their validation and quality systems, and the specialty sales force. Once they exist, adding a molecule costs far less than building the infrastructure from scratch. A rare-disease sales force already calling on the same specialists can carry Tepezza, Krystexxa and Uplizna at a low marginal cost, and that is largely how Amgen integrated Horizon.

Relative share is what matters, and Amgen's is large only against small biotechs. Measured against AbbVie, Merck, Pfizer, Johnson & Johnson or Novo Nordisk, Amgen is mid-sized; AbbVie's immunology business is several times the size of Amgen's inflammation franchise. Scale here deters startups without cowing the largest incumbents. Contract manufacturers have also lowered the capital needed to make biologics, which weakens the plant-based half of the advantage. Scale buys speed and better fixed-cost absorption. It does not set price.

Customer captivity: Moderate, and the payer holds the switch

Customer captivity: Moderate. Patients stable on a chronic biologic rarely switch voluntarily. Disease control is working, switching carries a risk of immunogenicity, and prescribers stick with what they know. Search costs are high for the patient and low for the specialist, who knows the alternatives well.

Three things blunt the lock. The economic decision belongs to the payer: formulary placement and rebate contracts determine which molecule is dispensed, and patient preference matters only where the payer allows a choice. Biosimilar substitution is precisely the moment the stickiness gets tested, and denosumab shows the result. And Amgen itself trades price for volume every year, which is not what a company with genuine pricing power does. Price sensitivity is high despite clinical inertia. The captivity is real but rented from payers, quarter by quarter.

Durability: the substitutes arrive on schedule

Biosimilars are the mechanical substitute: the same molecule at a lower price, admitted by the same regulator that once kept it out. In inflammation, the substitute is a different mechanism, including oral therapies that avoid injections. In pricing, Medicare negotiation substitutes for competition: negotiated prices for Enbrel take effect in 2026 and for Otezla in 2027, doing what a rival would have done, at a pace the company cannot influence.

The question that decides how long the moat lasts is whether Amgen can refill the portfolio faster than it empties. The five-year record is respectable on the top line, about 9% annual revenue growth, but since 2019 the company has spent roughly $41 billion on Otezla and Horizon. Growth has been bought as much as discovered. That is not disqualifying, since the acquired assets earn decent returns on operating capital, but it moves the test from science to capital allocation: can Amgen keep ROIC excluding goodwill above 20% without another large deal?

Three pieces of evidence would settle it. If denosumab erosion tracks Enbrel's slower curve instead of falling by a third, the patent cliff is shallower than feared. If Tepezza and Krystexxa stay uncontested, the rare-disease franchises hold pricing for longer than the inflammation business did. And if Repatha, Evenity, Tezspire and Blincyto keep growing at double-digit rates through 2027 on their own, the internal engine is doing the work. If ROIC including goodwill climbs back above 15%, the moat is wider than Moderate. If revenue growth requires another $20 billion acquisition, the pattern is a treadmill rather than a franchise.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Continue the comparison

Continue reading the full report

  1. Comparison and summaryHow Amgen Inc compares with its closest rivals, and our moat rating

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on Amgen Inc