West's stoppers and seals are written into the approved filings of the drugs they close, which makes the business an annuity. The question at $364.98 is not whether the annuity pays. It is how long it keeps compounding.
At $364.98, its close on October 2, West Pharmaceutical Services trades at 40.8 times the midpoint of its 2026 earnings guidance. If those earnings grow 12% a year for five years and the market still pays 30 times them in 2031, the shares offer about 5.5% a year. Ten percent needs 16.75% growth at that multiple, and the multiple is itself a forecast: paying 30 times in 2031 for a 10% return means expecting roughly two more decades of 12% growth. The business deserves a premium. Its components are written into the filings of the drugs they seal, and on every test the filings allow, it is winning. But guided 2026 earnings are only 4% above 2022's, growth is guided to slow in the second half, and one customer supplied at least $129.5 million of West's $180.9 million of growth in 2025. The price is a bet on duration, the one thing no filing discloses.
On October 1, Josh Brown and Sean Russo of Ritholtz Wealth Management put West on their list of the best stocks in the market, in a column for CNBC. Their case is the oldest one in markets, the one about selling shovels in a gold rush: biologics, vaccines, insulin and the new weight-loss drugs all have to be sealed in something, and West makes the stoppers, seals and plungers that do it. Brown needed seven words: "Whoever wins, West is selling to them."
Russo supplied the comeback. Customers spent the years after the pandemic working through components they had stockpiled, and in February 2025 West guided to adjusted earnings of $6.00 to $6.20 a share. It delivered $7.29, has raised its 2026 guidance twice, and now expects $8.85 to $9.05. The shares, as low as $223.83 within the past year, closed at $364.98.
The pitch is right about the business, and this note does not quarrel with it. It asks the narrower question every price asks: how much of this has already been paid for?
A stopper is a small plug of synthetic rubber, and the more valuable the drug behind it, the more is done to it: films and coatings, washing, sterilization, inspection by camera. West makes stoppers, plungers, seals, syringe and cartridge parts, and the devices patients use to inject themselves at home; a second segment, West Vantage, manufactures to customers' own designs. The premium components, which West calls High-Value Product Components, were 49% of second-quarter sales.
What turns the factory into a franchise is a sentence in West's annual report. Once a drug is approved with a West component, regulation makes it "difficult to change components and devices produced by one supplier" for another's, because the drug maker must generate a body of data showing that the substitute is equivalent and poses no new risk to patients. West concedes that if regulators eased that burden, competition would intensify. The moat is paperwork, and the paperwork belongs to the customer.
That is Adrian Slywotzky's installed base profit (The Profit Zone, 1997): thin on the first sale, rich on everything that follows. The installed base here is not a fleet of machines but a portfolio of approved drugs. The early work on a drug, the compatibility studies, the engineering support, the small lots for clinical trials, earns little; the profit comes after approval, when every dose consumes a component for as long as the drug sells in that form. The fit is exact in mechanism, with one gap: Slywotzky's test counts the units in use, and West does not report them. Every quarterly release since the middle of 2025 has cited the same "over 41 billion components and devices each year", a figure as steady as a stopper.
Written out, what the base spends with West each year is
where P is the number of patients taking drugs whose filings name a West component, d the injections each takes in a year, s West's share of those injections (whatever a second supplier has not taken), and a West's revenue per injection, which rises when a customer upgrades to a premium component or accepts a price increase. West is winning while
The filings cannot measure P, d or s. They do measure the two parts of a, price and mix, every quarter:
| Quarter | Price | Volume and mix | Premium share of components |
|---|---|---|---|
| Q3 2025 | +1.7 | +5.9 | 69.3 to 71.1 |
| Q4 2025 | +1.6 | +5.1 | 67.4 to 70.7 |
| Q1 2026 | +3.5 | +11.8 | 67.6 to 71.7 |
| Q2 2026 | +4.0 | +8.7 | 68.5 to 71.7 |
Price, and volume and mix: points of consolidated sales growth over the prior-year quarter, from the sales bridges in West's quarterly presentations, leaving out currency and the 2024 customer incentives that did not recur. Premium share: High-Value Product Components as a percent of those plus Standard Products, a year earlier and now.
On that evidence West is winning. Price rose in every quarter, and faster lately; customers kept climbing from standard to premium components; and volume and mix still added 7.8% to sales over the four quarters. Proprietary Products' gross margin reached 42.6% in the second quarter, up 2.5 points.
The test the filings cannot run is the one that matters most: how broad the base is. West's largest customer bought $485.9 million in 2025, 15.8% of sales; the largest a year earlier bought $356.4 million. Whether or not they are the same company, the top buyer added at least $129.5 million of the $180.9 million West grew, and GLP-1 drugs were 18% of second-quarter sales. The filings name no customer and tie none to a drug. But a base this concentrated is exposed in the two symbols the bull case takes for granted: d, if the next drugs are injected less often or taken as pills, two risks West's own annual report names; and s, if a second supplier is qualified. Datwyler, a competitor West names, used its 2026 half-year results to announce that it has begun supplying components for a leading GLP-1 weight-loss medicine from its plant in Middletown, Delaware. That is not a West loss. It is a reason to measure s.
No pattern in Slywotzky's companion book, Profit Patterns (1999), fits as exactly. The nearest, product to pyramid, describes West's ladder from standard to premium components, but West has run that ladder for years; what is new is how fast customers climb it, a change in a rather than a new pattern.
Underneath the profit model is the oldest mathematics in finance. Each approved drug is an annuity that pays West for as long as it sells in that form, and a portfolio of them is worth
where m is the share of F that becomes owner cash, r the return an owner requires, g the growth of F from the drugs that stay, and λ the share of the base lost each year: to cheaper presentations after a patent expires, to less frequent dosing or an oral form, or to a second supplier. At 40.8 times this year's earnings and a 10% return, the price needs g − λ of about 7.5% a year forever, before charging for the capital growth consumes. Nothing grows that fast forever, so the price must be paying for a long run of faster growth before a fade. The next section measures how long.
This reminds me of Ecclesiastes 10:1 (BSB): "As dead flies bring a stench to the perfumer's oil." West's sales come to at most about seven and a half cents for each component or device it ships, and a contaminated component can spoil a dose worth many times that. That asymmetry is the economics of the premium tier. What West sells, in the end, is the cost of keeping the flies out of the oil.
We rebuilt both models in the draft behind this note, reproduced every figure to the cent, and turned them around in the manner of expectations investing (Mauboussin, 2006): instead of asking what the shares are worth, which Vista does not publish, we ask what $364.98 offers a year on each set of assumptions. The first model pays the $0.88 dividend and sells the shares at a multiple of 2031 adjusted earnings:
| Earnings growth, 2027 to 2031 | 25 times | 30 times | 35 times | 40 times |
|---|---|---|---|---|
| 8% | −1.8% | 1.8% | 5.0% | 7.8% |
| 11.1%, the Research Desk's path | 1.0% | 4.7% | 8.0% | 10.9% |
| 12% | 1.8% | 5.5% | 8.8% | 11.8% |
| 15% | 4.5% | 8.4% | 11.7% | 14.7% |
| 18% | 7.2% | 11.2% | 14.6% | 17.7% |
| Growth a 10% return needs | 21.1% | 16.75% | 13.2% | 10.2% |
What $364.98 offers a year: the annual return at which five year-end dividends of $0.88 and a sale at the stated multiple of 2031 adjusted earnings equal the October 2 close, with growth compounding from the $8.95 guidance midpoint. The Research Desk's path, proposed while it challenged the draft, grows 12.5%, 12%, 11.5%, 10% and 9.5%.
The exit multiple is not a neutral assumption; it is a forecast about the forecast. A business expected to grow 3.5% a year forever while earning 20% on new capital is worth about 13 times earnings to an owner who wants 10%. For a buyer in 2031 to pay 30 times and still earn 10%, West must then be expected to grow 12% a year for about 22 more years, or 15% for 15. Thirty-five times asks for 27 more years at 12%, forty times for 32. Even if new capital earned 30%, 30 times would need 17 more years of 12% growth.
The second model works in cash: $8 a share in 2027, growing at a chosen rate for nine years and at 3.5% forever after.
| Cash growth, Years 2 to 10 | What $364.98 offers a year | Return on new capital it implies | Offers if new capital earns 20% |
|---|---|---|---|
| 12% | 7.5% | 59% | 7.2% |
| 15% | 8.4% | 74% | 7.8% |
| 18% | 9.3% | 88% | 8.4% |
| 20% | 10.0% | 98% | 8.7% |
Year 1 is 2027: 12% earnings growth on 71 million shares, plus $200 million of depreciation and amortization, less $300 million of capital spending and $45 million of working capital, is $7.98 a share, rounded to $8. The implied return on new capital is the growth rate divided by the fifth of earnings that build reinvests. The last column charges growth for its capital at 20%, about what West earned on all of its capital in 2025. Inputs and method are at the end of this note.
Three things stand out.
Year 1 asks for 30% more cash per share than West produced in the year to June: $8 against $6.15, which is trailing free cash flow of $436.9 million on the model's 71 million shares. The draft calls the $8 an assumption, and it is one.
Most of the value sits beyond 2036. At a 10% return, 63% to 70% of each cash case arrives after Year 10; on the 20% path, 2036 earnings are 5.8 times the 2026 midpoint.
The cash model's growth comes almost free of capital. Reinvesting a fifth of earnings while cash compounds at 12% to 20% implies a return on new capital of 59% to 98%. Price increases and spare capacity can do that for a while, not for a decade. Charged at 20%, each case offers 0.3 to 1.3 points less, a smaller gap than you might expect, because the decade's cash matters less than what follows it. In both models, the real assumption is duration.
This reminds me of Warren Buffett's letter to Berkshire Hathaway's shareholders for 1989: "Time is the friend of the wonderful business, the enemy of the mediocre." An installed base is a business whose value is mostly time. At 40.8 times earnings, the market has already decided which kind West is. What it has not been told is how much time it has bought.
West's guidance tells a story of acceleration, and its history one of recovery. Both are true. Management began 2026 expecting adjusted earnings of $7.85 to $8.20 a share and has raised the range twice, to $8.85 to $9.05. But adjusted earnings were $8.58 in 2022 and $8.08 in 2023, then fell to $6.75 in 2024, a year West spent waiting out its customers' destocking; the 2026 midpoint is 4.3% above 2022. The adjusted operating margin, 22.6% in the second quarter, is still below 2023's 23.4% and 2022's 26.4%. A recovery deserves a good multiple. It is not a decade of compounding.
Growth is also slowing. Sales grew 12.7% organically in the second quarter; the third-quarter guide is 7.0% to 8.9%, and 1.9% to 3.8% as reported, after the sale of the SmartDose 3.5mL injector to AbbVie and currency. Adjusted earnings rose 36.8% in the first half, and the full-year range leaves roughly 9% to 14% growth for the second ($4.35 to $4.55 a share, against $4.00). Growth stocks fall much further on a small disappointment than they rise on a beat of the same size (Skinner and Sloan, 2002).
The cash ran behind the earnings. First-half operating cash flow fell to $213.9 million from $306.5 million, and free cash flow to $128.0 million from $160.0 million, while net income rose to $292.8 million. The company cites working capital and incentive payments, and the balance sheet agrees: receivables rose $137.6 million from December on strong late-quarter sales, and accrued salaries fell $38.8 million. Measured crudely, receivables were 74.3 days of sales in June, against 69.1 a year earlier. Earnings that arrive as receivables persist less than those that arrive as cash (Sloan, 1996): a reason to watch collections, not a charge against West.
Meanwhile West spent $454.3 million on buybacks in the first half, 3.55 times its free cash flow, at an average of $258.03 a share; the October 2 close is 41% higher. Cash fell to $435.8 million from $791.3 million, against $203.0 million of debt due in July 2027 and an undrawn $500 million credit line.
Annex 1 is often told with the wrong calendar. Europe's revised rules for sterile manufacturing took effect on August 25, 2023, with one point postponed a year, and have been fully applicable since August 25, 2024, so no deadline lies ahead to pull demand forward. What remains is adoption. West counts about 6 billion components customers could upgrade; at the end of 2025 the Annex 1 projects it had completed represented about 15% of that, and it expects conversions to add about 2 points to 2026 growth. What each upgrade adds per component is not disclosed.
The rest of the news is mostly behind it. A cyberattack in May cost West Vantage about $7 million of second-quarter sales and, with cost inflation, cut its operating margin to 8.6% from 12.1%; the 10-Q says operations have fully recovered. In July West renewed for ten years its agreements with Daikyo Seiko, the 49%-owned Japanese company from which it licenses key premium technologies. On August 31 Michel Lagarde, formerly chief operating officer of Thermo Fisher Scientific, succeeded Eric Green as chief executive. The recovery predates him. The next chapter is his.
The literature is not hostile to West; it is hostile to long forecasts. Chan, Karceski and Lakonishok (2003) found sustained high growth in earnings rare, and no more common than chance would produce, and Fama and French (2000) found profitability reverting toward the average at roughly 38% a year. Both are averages across thousands of companies, and an installed base protected by its customers' own regulatory filings is exactly the kind of business that might beat them. That is the bull case. It is also what the price already assumes.
Here is the best case we can build without cheating. The Annex 1 conversion was about 15% done at the end of 2025. Biologics, 43% of sales, grew 29.2% organically in the second quarter, and premium components outside GLP-1 grew at a high-teens rate. Price added 3.5% and 4.0% to sales in the last two quarters, and management has a recent habit of guiding low. The first-half cash shortfall is timing, as the company says, and reverses. Daikyo is secured for ten years, and the new chief executive once ran Thermo Fisher's pharmaceutical services business.
On that case earnings grow 15% a year for five years, the market pays 35 times them in 2031, and $364.98 offers 11.7% a year. Eighteen percent growth at 30 times offers 11.2%; cash growing 20% a year for a decade, 10.0%.
That is a good return, and each assumption can be defended on its own. Together they ask West to do what the record says is rare: grow fast for a decade, then persuade the buyer at the end that it will do so for another. An investor who believes the installed base can do that is offered 10% to 12% a year. One who expects West to grow like a very good business rather than a rare one, 12% for five years and 30 times at the end, is offered about 5.5%. Which of them is right is the reader's decision, not ours.
Likely, we can get some of it from primary research, which is what Vista is all about. The filings will report, a quarter at a time, whether price and mix keep rising. People who qualified, bought and made these components in earlier cycles can tell us now how long such conversions run, and what makes a drug maker add a second supplier.
We would askAs a matter of past practice, what did it take, in data, months and money, to qualify a second stopper or plunger supplier for a commercial drug, and what usually triggered the decision?
The answer that would change the view"Large customers dual-source as a matter of policy within a few years of launch."
We would askHow were component contracts generally structured: volume commitments, price escalators, rebates? When a supplier raised prices 3% to 4% a year, how long did that last before the next negotiation reset it?
The answer that would change the view"Increases like that are catch-ups, and they reset at renewal."
We would askWhich upgrades permanently raise the component content of each dose, and which were one-time changes? In general, how far through its conversion was the industry by the end of 2025?
The answer that would change the view"The large conversions are done. What remains is small and slow."
We would askAs next-generation GLP-1 and biologic drugs move to less frequent dosing or new presentations, what happens to component content per patient per year?
The answer that would change the view"Monthly dosing removes more injections than premium components add revenue to each one."
Every conversation would draw on past, general experience only: no current employees of West or of its customers and suppliers, nothing confidential, and every Advisor screened for conflicts before a word is said. For a client engagement, Vista would combine this note with those interviews: two or three Advisors, structured conversations, and one written brief that says where they agree, where they split, and what would change the answer.
We will score this note in February 2027, after West reports its fourth quarter and full year 2026 and gives its guidance for 2027 (it reported on February 13 in 2025 and on February 12 in 2026), and publish the result on the scorecard whether it flatters us or not. The tests are fixed today, so they cannot drift:
Four or five passes, and the annuity is compounding at the pace the price needs, for now. Two or fewer, and the price is running ahead of the business. Three is too early to say. We will also report what the price then offers on each case, and what share of 2026 sales the largest customer took, from the annual report.
West sells permanence by the dose. At $364.98, the market is buying it by the decade.
Both models are the draft's, rebuilt in code and reproduced to the cent before anything was turned around. The five-year model: value = five year-end dividends of $0.88 + exit multiple × 2031 adjusted earnings per share, discounted at the required return, with 2026 adjusted earnings of $8.95 (the guidance midpoint) as time zero, 2027 as Year 1 and 2031 as Year 5; the horizon runs from October 2, 2026 to autumn 2031, a small calendar mismatch the model accepts. The ten-year model: Year-1 cash of $8 a share, growing at the stated rate in Years 2 to 10 and at 3.5% a year after; Year 1 (2027) is discounted a full year from October 2, 2026, although its cash arrives through 2027. Stock compensation stays an expense in both. The return each case offers is the discount rate at which its dividends and sale, or its cash, equal the $364.98 close. The return on new capital a cash case implies is its growth rate divided by the 20.37% of earnings the Year-1 build reinvests ($145 million of $711.7 million). The 20% columns charge growth for capital: cash = earnings × (1 − g / 0.20), with earnings growing from $10.02 a share, and 3.5% growth at the same 20% after Year 10. The exit-multiple test values a business that grows at the stated rate for n years and 3.5% after, paying out whatever growth at 20% on new capital does not need, at a 10% return. West's 2025 return on capital, about 20%: adjusted operating profit of $622.4 million after the 19.5% adjusted tax rate, over average equity plus debt less cash. The 7.5% a year of growth, net of the base lost, that the multiple implies forever is 10% less the earnings yield on guidance: 8.95 / 364.98 = 2.45%. Every figure in this note was recomputed in code from these inputs and West's filings.
| Input | Five-year model | Ten-year model |
|---|---|---|
| Starting point | 2026 EPS $8.95 | 2027 cash $8.00 a share |
| Growth | 8% to 18% a year, 2027 to 2031 | 12% to 20% a year, Years 2 to 10 |
| After the horizon | Sale at 25 to 40 times 2031 EPS | 3.5% a year forever |
| Dividends | $0.88 a year, year-end | Within the cash |
| Year-1 build | n/a | EPS $10.024 × 71m shares + $200m D&A − $300m capex − $45m working capital |
| Required return | 10%, and solved for | 10%, and solved for |
This note is research, not investment advice. It states what a market price assumes under labeled assumptions; it is not a recommendation to buy, sell or hold any security, and the decision belongs to the reader. As of October 3, 2026, Russ Rosenzweig, Vista's founder, owns no West Pharmaceutical Services shares. The note began as a research question Russ Rosenzweig put to the Vista Research Desk, which challenged his draft over four rounds and then reviewed the full manuscript; the models and every figure were then rebuilt and checked against the company's filings.