Pfizer PFE
Healthcare · Drug Manufacturers - General · Synthos Deep Dive · 2026-08-04
$27.96
Buy — Tactical
—
—
The Overview
Pfizer makes prescription medicines and vaccines. It is one of the largest drug companies in the world and it reported its latest quarter today.
The results contained two numbers that appear to contradict each other. On the measure analysts follow, Pfizer earned 77 cents a share against an expected 68 cents — a 13% beat — on sales of $15.0 billion, about 4% more than expected. On the official accounting measure, it lost four cents a share.
Both are true. The difference is a single accounting line — "other deductions" — which was $3.7 billion this quarter against $0.7 billion a year ago. That $3 billion swing wiped out the $4.7 billion the business earned from operations. The filing does not say what it consists of, so neither do we.
Underneath, the business is fine: sales grew 2.6%, and the money it earns from drugs it markets jointly with partners grew 19%. It spent $2.8 billion on research in three months, more than a fifth of its revenue.
The reason to own the shares is the dividend. At $25.41 they pay $1.72 a year — a 6.77% yield, by far the highest of the twelve companies in this group and roughly double what a US government bond pays. The shares also barely move with the market: their beta of 0.28 is the lowest here by more than half.
The reason to be careful is that analysts expect Pfizer's profits to fall. Consensus has earnings going from $2.94 a share this year to $2.44 by 2028, and revenue falling 13.5% by 2029. That is the "patent cliff" — several of its biggest drugs lose their exclusive rights and cheap copies arrive. This is not a hidden risk; it is what the analysts are already forecasting, which is why the shares cost only about 8.6 times this year's expected earnings.
The number that decides everything is whether the dividend is safe. It costs about $9.8 billion a year. The company generated about $11.0 billion of spare cash over the past year. That is a margin of only 12% — the tightest of any company in this group. On the profit measure the cover is more comfortable, at 1.77 times, and would still be 1.34 times even at the lowest year analysts forecast.
Two further notes. The chief financial officer leaves on 15 August, eleven days from now, for a job in consumer goods; the company states it is unrelated to results. And our research library has an attribution problem on this name: six of the 28 expert claims about Pfizer are filed under a channel named for the chief executive of a rival company, Eli Lilly. We have excluded all six.
Our estimate of fair value is $28, about 10% higher, plus the 6.8% dividend. The average analyst says $26.88 — and the analyst community as a whole rates it a Hold.
- Downside Risk 5/10. The steadiest share price here and a corrected balance sheet that is much better than it screens — against a shrinking earnings base and thin cash cover on the dividend.
- Growth Quality 3/10. Consensus forecasts decline. That is the honest score.
- Exponential Potential 3/10. It spends more on research than anyone here, and nothing available to us lets us value what it buys.
Putting a number on it: our fair-value estimate is $28 against a current price of $27.96 — real upside if our numbers are right.
Our summary metrics
Downside Risk (lower = safer)5/10Moderate
"Rated 5 — the lowest-volatility security in this batch attached to a genuinely declining earnings base and a dividend with a thin cash-flow margin. Supports: beta of 0.28, less than half the next-lowest name here; a trailing free-cash-flow yield of 7.59% against a 6.77% dividend yield; the dividend covered 1.77 times by trailing adjusted earnings of $3.05; interest coverage of 5.92x; corrected funded net debt of $51.529 billion at roughly 2.6x our constructed adjusted EBITDA of about $20.9 billion, against the vendor's alarming 4.93x; research and development spending of 20.1% of revenue, which is real reinvestment rather than harvest; a second-quarter beat of 12.9% on adjusted EPS and 4.4% on revenue; and a price only 8.0% above the 52-week low. Against that: free cash flow covers the dividend by only about 1.12 times, which is the thinnest margin anywhere in this batch and is the single number on which the thesis turns; goodwill of $71.264 billion plus intangibles of $53.731 billion is $124.995 billion against total equity of $86.476 billion, so tangible equity is roughly NEGATIVE $38.5 billion; GAAP results are unusable, with the June quarter posting a $248 million net loss on $4.7 billion of operating income because 'Other (income)/deductions—net' was $3.716 billion; consensus models EPS falling from $2.943 in 2026 to $2.438 in 2028; the chief financial officer, Dave Denton, notified the company on 2026-06-16 that he steps down effective 2026-08-15 — ELEVEN DAYS after this dive; and the sell side is a Hold, 15 buy against 23 hold and 1 sell, with a low target of $25 below spot. Not a 4, because the earnings decline is real and the dividend cover is thin. Not a 6, because the balance sheet is not stretched once corrected and the beta is 0.28."
Growth Quality3/10Low
"Rated 3 — this is a shrinking business on the consensus path and the file is honest about it. Consensus revenue: $61,775M in 2026, $59,294M in 2027 (−4.0%), $54,738M in 2028 (−7.7%), $53,406M in 2029 and $53,804M in 2030 — a 13.5% decline from 2026 to 2029 before a flat 2030. Consensus EPS falls in step: $2.943, $2.835, $2.438, $2.304, $2.350. That is the loss-of-exclusivity cliff and 18 to 21 analysts model it. What is growing: the June quarter delivered total revenues of $15,034M, up 2.6%, and $29,484M for the half, up 3.9% — so the decline has not started yet. Within it, alliance revenues grew 18.7% to $2,697M and royalty revenues 11.3% to $474M, while product revenues FELL 0.8% to $11,863M — the growth is coming from partnered assets rather than owned ones. Research and development rose 13.2% to $2,809M, 18.7% of quarterly revenue and 20.1% on a trailing basis, which is the highest reinvestment rate in this batch and is the company buying its way past the cliff. Business development is the visible mechanism: our knowledge base records $10B+ of deployable capacity, a $1.25B upfront payment to license 3SBio's SSGJ-707, and the contested acquisition of Metsera. A 3: current trading is fine, the forecast is a decline, and everything above a 3 would have to be bought."
Exponential Potential3/10Low
"Rated 3 — a large, cash-generative pharmaceutical whose only route to non-linearity is a pipeline nobody in this file can value. What is real: research and development spending of $11,976M in 2025 and 20.1% of trailing revenue — the highest reinvestment ratio of the twelve names here — plus a business-development programme our knowledge base describes at conviction 70 as '10-15 deals, higher next year, focused on internal medicine and I&I rather than vaccines or large oncology', with $10B+ of capacity at conviction 75. Two specific assets appear: 3SBio's SSGJ-707, licensed for $1.25B upfront, with data described at conviction 60 as '~67.5% ORR, near-100% disease control, deepening responses—among best in class'; and Metsera, which Pfizer bid $7.3B for and, after a roughly $9B counterbid from Novo Nordisk and an FTC threat over the deal structure, 'aggressively won... to find post-pandemic footing.' Against that: the vendor payload contains no pipeline data of any kind, `seg_prod` and `seg_geo` are both empty, and no source in the data contract lets us assess a single asset's probability of success. The one thing that IS quantified points the other way — our knowledge base records that Pfizer itself put the Inflation Reduction Act Part D redesign at roughly 1.6% of the top line. A 3: the optionality exists, it is being paid for out of cash flow, and nothing in the sources available makes it measurable."
Fair value$28 $19–$35
What does “fair value” mean?
Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.
The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, plus an independent, public-math cross-check further down the page (“Check our number”) so you can judge it for yourself.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Buy — CoreOwn it as a foundation — start or add now, size it for years, let dips be gifts.
Buy — TacticalGood price + confirmed trend + a defined exit — buy the setup, not a marriage.
WatchWe want the business, just not at this price/setup — act only when the listed trigger hits.
HoldFine to keep if you own it — no reason to buy more; new money does better elsewhere.
AvoidDon't own it — the problem is the business or the expectations, so a cheaper price won't fix it.
The Road Ahead
What we expect to matter in each window, and the evidence that would prove us wrong.
Short term 0-6 months
Tailwind- Driver
- "The quarter is already reported and it beat on both lines. Adjusted diluted EPS of $0.77 against a $0.682 consensus is a 12.9% beat, and revenue of $15,034M against $14,399M is 4.4% — the fifth consecutive beat in the file and the fifth by a wide margin (+34.5%, +37.4%, +16.4%, +3.9%, +12.9%). The market took it constructively: the shares closed at $25.41, up 1.52% on 55.4 million shares — the second-heaviest volume in this batch — having opened at $25.04 and traded to $25.64. The chart is neutral-to-improving: 1.4% above a 50-day moving average of $25.06 and 1.6% below a 200-day of $25.84, with RSI at 59.1 and MACD marginally positive at +0.08, sitting 11.0% below the 52-week high of $28.55 and 8.0% above the low of $23.53. Twelve-month return is +8.2% against SPY's +24.3% — behind, but the only positive twelve-month number among the four worst relative performers in this batch. And you are paid 6.77% to wait. The near-term risk that is NOT in the price is the chief financial officer's departure on 2026-08-15."
- What we’re watching
- "The composition of 'Other (income)/deductions—net', which was $3,716M in the June quarter against $739M a year earlier and turned $4.7 billion of operating income into a pre-tax loss. The extracted 10-Q does not disaggregate it and the next disclosure that does is the material item to read. The chief financial officer transition on 2026-08-15 — Dave Denton's notice explicitly states the departure 'is not related to the Company's financial or operating results or to any disagreements', which is the standard formulation and is recorded as such rather than discounted. Whether product revenues, down 0.8% in the quarter, return to growth, since alliance and royalty revenues are doing all the work. Research and development at 20.1% of revenue and rising 13.2% year on year, which is the cost of the pipeline and the swing factor in near-term adjusted earnings. And the 2026-11-03 print against $0.817 of adjusted EPS on $16,153.4M of revenue."
- Confidence
- Medium
Medium term 6-24 months
No differentiated view- Driver
- "The medium term is the loss-of-exclusivity cliff and consensus has already written it down. Revenue estimates fall from $61,775M in 2026 to $59,294M in 2027 and $54,738M in 2028 — a 11.4% decline across two years on 18 and 21 analysts — and EPS from $2.943 to $2.835 to $2.438. This is not a risk to the thesis; it IS the thesis, and the question is only whether the price already contains it. At 8.6x the 2026 consensus and 10.4x the 2028 consensus, the multiple RISES as earnings fall, which is the market's way of saying it is valuing something beyond the cliff. What Pfizer is doing about it is visible: research and development of $11,976M in 2025 and 20.1% of trailing revenue; a business-development programme our knowledge base puts at 10-15 deals a year with $10B+ of capacity; a $1.25B upfront licence of 3SBio's SSGJ-707; and the contested acquisition of Metsera, won against a roughly $9B counterbid from Novo Nordisk. Two quantified headwinds sit alongside: the Inflation Reduction Act Part D redesign, which Pfizer itself put at roughly 1.6% of the top line, and a payer-liability shift on catastrophic coverage from 15% to 60% that our knowledge base describes at conviction 70 as 'driving outright payer nos on expensive specialty drugs.'"
- What we’re watching
- "Dividend cover, which is the number that decides this security. The trailing dividend of $1.72 a share on 5,699 million shares costs roughly $9.8 billion a year; trailing free cash flow implied by the vendor's 7.59% yield is roughly $11.0 billion. That is 1.12 times cover — the thinnest margin in this batch — against 1.77 times on adjusted earnings. If consensus EPS of $2.304 for 2029 is right, the payout ratio at the trough is 74.7% of adjusted earnings, which is survivable; if free cash flow falls faster than earnings, it is not. Whether product revenues stabilise; they fell 0.8% in the June quarter while alliance revenues rose 18.7%. Whether business-development spending converts — Metsera and SSGJ-707 are the two named assets and neither has disclosed economics in this archive. Whether the Part D redesign's roughly 1.6% top-line drag proves to be the whole of the Inflation Reduction Act effect or the start of it. And whether the balance sheet stays where it is: corrected funded net debt of $51.5 billion at roughly 2.6x adjusted EBITDA is comfortable, but a large acquisition funded with debt into a declining earnings base would not be."
- Confidence
- Medium
Long term 2+ years
Neutral- Driver
- "Beyond the cliff the question is whether Pfizer's reinvestment rate buys a business as large as the one it is losing, and nothing in the sources available to this dive answers it. What can be said: the company spends more of its revenue on research and development than any other name in this batch, 20.1%; it is buying externally at scale, with our knowledge base recording $10B+ of deployable capacity and a shift 'from small add-ons to larger acquisitions'; and it has demonstrated it will pay up, winning Metsera against a competing bid roughly 19% higher per share and an FTC challenge to the deal structure. The sector backdrop from our knowledge base is mixed but currently favourable to the shares rather than to the science: biotech_hangout (2026-01-16, conviction 68) records that 'generalists are rotating into pharma/biotech because it looks cheap versus an expensive AI-driven S&P; interest is broadening beyond mega-caps to Merck, Bristol, Pfizer, Gilead, Amgen' — which is a flows argument, not a fundamental one, and is treated as such. The structural bear point is all_in (2026-01-15, conviction 80): 'US is losing the biotech race to China (5% to 42% of licensing deals since 2022); has ~3-year window to reform or fall behind on speed' — a claim Pfizer's own $1.25B licence of a Chinese asset, 3SBio's SSGJ-707, simultaneously illustrates and mitigates."
- What we’re watching
- "Whether the pipeline replaces the cliff. No source in the data contract lets us assess a single asset — the vendor payload contains no pipeline data, and `seg_prod` and `seg_geo` are both empty — so this must be watched through revenue outcomes rather than through probability estimates. Whether the dividend survives a trough year; at consensus 2029 EPS of $2.304 the payout ratio is 74.7% of adjusted earnings and free cash flow cover is the binding constraint. Whether the Inflation Reduction Act's effects broaden beyond the roughly 1.6% top-line drag Pfizer has quantified. Whether Chinese licensing shifts from opportunity to competition — Pfizer is currently a buyer of Chinese assets, which works only while it is the buyer. Governance and continuity: Albert Bourla has led the company since 2019 and the chief financial officer departs on 2026-08-15; the file contains no chief-executive succession disclosure. And the reputational tail our knowledge base records at conviction 80 from a named third-party source — the 2010 federal fraud and racketeering conviction — which is history but is the kind of history that recurs in this sector."
- Confidence
- Low
Exponential Potential
Exponential Potential3/10Low
"Rated 3 — a large, cash-generative pharmaceutical whose only route to non-linearity is a pipeline nobody in this file can value. What is real: research and development spending of $11,976M in 2025 and 20.1% of trailing revenue — the highest reinvestment ratio of the twelve names here — plus a business-development programme our knowledge base describes at conviction 70 as '10-15 deals, higher next year, focused on internal medicine and I&I rather than vaccines or large oncology', with $10B+ of capacity at conviction 75. Two specific assets appear: 3SBio's SSGJ-707, licensed for $1.25B upfront, with data described at conviction 60 as '~67.5% ORR, near-100% disease control, deepening responses—among best in class'; and Metsera, which Pfizer bid $7.3B for and, after a roughly $9B counterbid from Novo Nordisk and an FTC threat over the deal structure, 'aggressively won... to find post-pandemic footing.' Against that: the vendor payload contains no pipeline data of any kind, `seg_prod` and `seg_geo` are both empty, and no source in the data contract lets us assess a single asset's probability of success. The one thing that IS quantified points the other way — our knowledge base records that Pfizer itself put the Inflation Reduction Act Part D redesign at roughly 1.6% of the top line. A 3: the optionality exists, it is being paid for out of cash flow, and nothing in the sources available makes it measurable."
What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.
Clinical pipeline
Clinical-trial data for PFE hasn’t been pulled yet — check back soon.
Deeper analysis
Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.
> ### Results released today — 2026-08-04
> Pfizer announced second-quarter 2026 results on 2026-08-04 and filed the 10-Q the same day. The actuals ARE captured in the payload:
> Adjusted diluted EPS $0.77 against a $0.682 consensus — a 12.9% BEAT. Revenue $15,034M against $14,399M — a 4.4% BEAT.
> GAAP diluted EPS was MINUS $0.04, against +$0.51 a year earlier.
> The entire $0.81 gap sits in one line: "Other (income)/deductions—net" of $3,716M against $739M — a $2,977M swing that turned $4.7 billion of operating income into a $653 million pre-tax loss. The extracted filing does not disaggregate it and we do not estimate around it.
> The shares closed +1.52% at $25.41 on 55.4 million shares.
Reference table
| Street consensus | $26.88 (+5.8%) · median $27.00 (+6.3%) · high $30 · low $25 — 1.6% BELOW spot · 15 buy / 23 hold / 1 sell across 39 analysts, consensus "Hold" |
| Valuation | 8.33x TTM ADJUSTED EPS ($3.05) · 8.63x 2026E · 8.96x 2027E · 10.42x 2028E — the multiple RISES as earnings fall · ~9.4x adjusted EBITDA on rebuilt EV · vendor priceToEarningsRatioTTM of 33.58x REJECTED |
| The dividend, and the number that decides this | $1.72 trailing, a 6.77% yield. Covered 1.77x by trailing adjusted EPS of $3.05, 2.26x GAAP EPS the other way (the payout is 226% of GAAP), and — critically — only about 1.12x by trailing free cash flow (7.59% free-cash-flow yield against a 6.77% dividend yield) |
| The quarter filed today | Total revenues $15,034M (+2.6%): product $11,863M (−0.8%), alliance $2,697M (+18.7%), royalty $474M (+11.3%) · R&D $2,809M (+13.2%) · Restructuring $457M (vs −$18M) · Other (income)/deductions—net $3,716M (vs $739M) · Pre-tax −$653M · Net loss attributable −$248M · GAAP diluted EPS −$0.04 |
| Consensus models a decline — this is the thesis, not a risk to it | Revenue $61,775M (2026) → $59,294M (2027) → $54,738M (2028) → $53,406M (2029), a 13.5% erosion. EPS $2.943 → $2.835 → $2.438 → $2.304 |
| Leverage — the vendor overstates it badly | Vendor netDebt $66,274M and netDebtToEBITDATTM 4.93x. The figure ignores $12,454M of short-term investments, and the EBITDA denominator is depressed by the charges above. Corrected funded net debt $51,529M at roughly 2.6x our constructed adjusted EBITDA of ~$20.9B |
| Unscheduled event 11 days out | 8-K, 2026-06-18: chief financial officer Dave Denton steps down effective 2026-08-15 for a role "outside of the pharmaceutical industry in consumer goods", with the standard statement that it is "not related to the Company's financial or operating results or to any disagreements" |
| Conviction | Low — 28 claims, and the lane requires two exclusions before use. SIX claims come from a channel slug named for the chief executive of ELI LILLY and are about PFIZER — the documented cross-company attribution failure, confirmed. Two more are Albert Bourla speaking with no management tag |
| Technicals | −11.0% from the 52-week high of $28.55, +8.0% above the low of $23.53; +1.4% above the 50-DMA, −1.6% below the 200-DMA; RSI 59.1; MACD +0.08; 12-month +8.2% vs SPY +24.3%; six-year maximum drawdown −58.5% |
What the experts actually said 2 traceable claims on PFE · showing the highest-conviction voices
“Biotech is extremely interesting — mRNA firms like Moderna and Pfizer had the vaccine blueprint within weeks of sequencing; many more drugs and innovations are coming beyond obesity/GLP-1.”
“On valuation Pfizer or Merck look preferable to Lilly — lower P/E (Lilly ~70x), higher dividend yield, and higher net margin/ROA.”
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
Price & moving averages 12 months · 50 & 200-day averages · 52-week range
Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.
Data summary: last close $27.96, 9% above the 50-day average ($26), 7% above the 200-day average ($26) — an uptrend. 2% below the 52-week high of $29, 18% above the 52-week low of $24.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Data summary: price $27.96 is currently inside the band (band $25–$29).
RSI (14) momentum gauge · 0–100
Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 63.
MACD 12 / 26 / 9 · trend & momentum
The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.
Data summary: MACD is currently above its signal line by 0.05, positive momentum.
Relative performance vs S&P 500 & its sector (XLV (sector)), set to 100 a year ago
Solid = PFE · dashed = S&P 500 · dotted = XLV (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
Price$27.96
Market cap$159B
P/E trailing37×
P/E FY26E / FY27E9× / 10×
EV / Sales3.3×*
EV / EBITDA16.9×*
Gross margin71.3%
Net margin6.8%
Dividend yield6.15%
Beta0.28
52-wk range$24 – $29
RSI(14)60
50 / 200-DMA$26 / $26
12-mo return+12% (SPY +19%)
Street target$28 ($26–$31)
Analyst grades15 Buy · 23 Hold · 1 Sell
FMP ratingB-
Next earnings'2026-11-03 (third-quarter 2026 earnings, 91 days away; vendor consensus adjusted EPS $0.817 on revenue of $16,153.4M). SECOND-QUARTER RESULTS WERE RELEASED ON THE DIVE DATE, 2026-08-04, with the 10-Q filed the same day and the actuals already captured — adjusted EPS $0.77 against $0.682 (a 12.9% beat) and revenue $15,034M against $14,399M (a 4.4% beat). A nearer, unscheduled event: chief financial officer Dave Denton steps down effective 2026-08-15, eleven days after this dive.'
1. The quarter filed today, line by line
From the 10-Q filed 2026-08-04 (US$M except per share):
| Q2 2026 | Q2 2025 | Change | H1 2026 | H1 2025 | Change | |
|---|---|---|---|---|---|---|
| Product revenues | $11,863 | $11,954 | −0.8% | $23,578 | $23,248 | +1.4% |
| Alliance revenues | $2,697 | $2,273 | +18.7% | $5,036 | $4,386 | +14.8% |
| Royalty revenues | $474 | $426 | +11.3% | $870 | $734 | +18.5% |
| Total revenues | $15,034 | $14,653 | +2.6% | $29,484 | $28,367 | +3.9% |
| Cost of sales | $4,092 | $3,778 | +8.3% | $7,640 | $6,624 | +15.3% |
| Selling, informational and administrative | $3,411 | $3,415 | −0.1% | $6,372 | $6,446 | −1.1% |
| Research and development | $2,809 | $2,482 | +13.2% | $5,299 | $4,685 | +13.1% |
| Acquired in-process R&D | $16 | $2 | — | $153 | $11 | — |
| Amortisation of intangible assets | $1,185 | $1,211 | −2.1% | $2,368 | $2,421 | −2.2% |
| Restructuring and acquisition-related costs | $457 | ($18) | — | $557 | $660 | −15.6% |
| Other (income)/deductions—net | $3,716 | $739 | +$2,977M | $4,577 | $1,692 | +$2,885M |
| Income/(loss) before taxes | ($653) | $3,044 | — | $2,517 | $5,828 | −56.8% |
| Provision/(benefit) for taxes | ($407) | $141 | — | $54 | ($48) | — |
| Net income/(loss) attributable to Pfizer | ($248) | $2,910 | — | $2,440 | $5,877 | −58.5% |
| GAAP diluted EPS | ($0.04) | $0.51 | — | $0.43 | $1.03 | −58.3% |
| Adjusted diluted EPS (earnings calendar) | $0.77 | $0.78 | −1.3% | $1.52 | $1.53 | −0.7% |
| Weighted-average diluted shares | 5,699 | 5,706 | −0.1% | 5,733 | 5,708 | +0.4% |
Four readings.
First, the loss is one line. Operating performance — revenue less cost of sales, selling costs, research and development and amortisation — produced roughly $3.5 billion of income before restructuring. "Other (income)/deductions—net" of $3,716 million took it below zero. A year earlier the same line was $739 million. The extracted 10-Q does not disaggregate the $3,716 million, and we do not guess at it. What we can say is that the swing is $2,977 million, that it is the difference between a $0.77 adjusted quarter and a minus-$0.04 GAAP quarter, and that an $0.81-per-share reconciling item is 105% of the adjusted figure.
Second, the growth is coming from partners, not products. Product revenues FELL 0.8% to $11,863 million while alliance revenues rose 18.7% to $2,697 million and royalties 11.3%. Alliance and royalty revenue together is now 21.1% of the total, up from 18.4%. That is a company whose owned portfolio is flat and whose partnered portfolio is growing — which is exactly what a business managing a patent cliff through business development looks like, and it is not visible anywhere in the vendor payload.
Third, research and development is rising into the cliff, not falling. $2,809 million in the quarter, up 13.2%; 18.7% of quarterly revenue and 20.1% on a trailing basis. Selling, informational and administrative expense was flat. The company is holding commercial cost and increasing research spend — the correct behaviour, and the reason near-term adjusted earnings are not higher.
Fourth, restructuring turned from a credit to a charge. $457 million in the quarter against a $18 million CREDIT a year earlier, and $557 million for the half. A cost-reduction programme is running.
2. The forecast is a decline, and that is the whole valuation question
Consensus estimates, from the est block (US$M except per share):
| 2026E | 2027E | 2028E | 2029E | 2030E | |
|---|---|---|---|---|---|
| Revenue | $61,774.8 (18 analysts) | $59,293.6 (18) | $54,737.8 (21) | $53,405.9 (10) | $53,803.8 (10) |
| Revenue growth | −1.3% (vs 2025 actual) | −4.0% | −7.7% | −2.4% | +0.7% |
| Adjusted EPS | $2.943 (14) | $2.835 (13) | $2.438 (6) | $2.304 (4) | $2.350 (2) |
| EPS growth | — | −3.7% | −14.0% | −5.5% | +2.0% |
| P/E at $25.41 | 8.63x | 8.96x | 10.42x | 11.03x | 10.81x |
| Dividend cover (on $1.72) | 1.71x | 1.65x | 1.42x | 1.34x | 1.37x |
The multiple rises every year because earnings fall every year. That is the correct way to read this security, and it is the opposite of every other name in this batch.
Estimate coverage is adequate on 2026, 2027 and 2028 revenue (18, 18 and 21 analysts) and progressively thin on EPS — 14, 13, SIX, FOUR and TWO. The 2028, 2029 and 2030 EPS rows rest on six, four and two analysts respectively and are used only as directional cross-checks, never as anchors. Note also the 2028 EPS dispersion: $1.578 low against $2.780 high on six analysts, a range equal to 49% of the mean — genuine disagreement about how deep the cliff goes.
The 2023 estimate row reports netIncomeAvg of MINUS $2,593M and epsAvg of MINUS $0.459 against an actual 2023 net income of $2,133M and EPS of $0.38. We do not use it, and note that a consensus that expected a loss where the company reported a profit is a reminder of how wide the error bars on this business have been.
est.ebitdaAvg and est.ebitAvg are not used. The ratios are effectively fixed — ebitdaAvg runs 29.49%, 29.49%, 29.49%, 29.49% and 29.49% of revenueAvg across 2026 to 2030, and ebitAvg 21.32% throughout — which is the fixed-ratio signature. Independently, the 2025 ebitAvg of $13,202.7M against actual operating income of $15,437M is 14.5% low. All forward valuation runs on epsAvg and revenueAvg.
3. Balance sheet — the vendor's leverage figure is badly overstated
At 2025-12-31 (the payload's most recent annual balance sheet; US$M):
| 2025-12-31 | 2024-12-31 | |
|---|---|---|
| Cash and cash equivalents | $1,142 | $1,043 |
| Short-term investments | $12,454 | $19,434 |
| Cash and short-term investments | $13,596 | $20,477 |
| Inventory | $10,654 | $10,851 |
| Goodwill | $71,264 | $68,527 |
| Intangible assets | $53,731 | $55,412 |
| Total assets | $208,160 | $213,396 |
| Short-term debt | $3,484 | $6,946 |
| Long-term debt | $61,641 | $57,405 |
| Lease obligations (vendor field) | $2,291 | $2,642 |
Vendor totalDebt | $67,416 | $66,993 |
Vendor netDebt | $66,274 | $65,950 |
| Corrected net debt (total debt less cash AND short-term investments) | $53,820 | $46,516 |
| Corrected FUNDED net debt (excluding lease obligations) | $51,529 | $43,874 |
| Total stockholders' equity | $86,476 | $88,203 |
> The vendor's netDebt of $66,274M is total debt of $67,416M less cash of $1,142M ONLY. It ignores $12,454M of short-term investments — the standing omission from the defect register, and here it is the largest single dollar error in this batch's balance sheets.
And the leverage ratio compounds the error. netDebtToEBITDATTM reads 4.93x, which implies a trailing EBITDA of about $13,447M. That denominator is depressed by exactly the charges described in Section 1 — the vendor's own quarterly ebitda series reads $5,648M, $756M, $5,021M and $1,199M across the last four quarters, with two of the four crushed by the "other deductions" line.
Our construction, method stated. Trailing operating income from the vendor's quarterly series is $3,550M + $3,688M + $4,032M + $4,706M = $15,976M. Adding back annual depreciation and amortisation of approximately $4,900M gives adjusted EBITDA of roughly $20,876M.
| Vendor | Corrected | |
|---|---|---|
| Net debt | $66,274M | $51,529M (funded) / $53,820M (incl. leases) |
| Trailing EBITDA | $13,447M (implied) | ~$20,876M (adjusted) |
| Net debt / EBITDA | 4.93x | ~2.6x |
| Enterprise value | $207,041M | ~$196,352M |
| EV / adjusted EBITDA | 16.40x | ~9.4x |
The vendor describes a company at nearly five turns of leverage on 16x EBITDA. The filings and the corrected arithmetic describe one at roughly 2.6 turns on 9.4x. No greater precision is claimed for the adjusted EBITDA figure than the method supports, and both versions are shown.
Two genuine balance-sheet observations that are not artefacts. Goodwill of $71,264M plus intangibles of $53,731M is $124,995M against total equity of $86,476M — tangible equity of roughly NEGATIVE $38.5 billion, which for a company facing patent expiries on the assets those intangibles represent is a real impairment risk rather than an accounting curiosity. And short-term investments fell from $19,434M to $12,454M in 2025, a $6,980M reduction — cash being deployed, and the file does not say into what.
4. The dividend — the number the whole thesis turns on
dividendPerShareTTM is $1.72, a 6.77% yield. profile.lastDiv is null, and no filing in this archive states the current declared quarterly rate, so no forward run-rate is asserted — the trailing figure is used throughout and labelled as such.
Three cover measures, and they disagree in an informative way:
| Measure | Calculation | Cover |
|---|---|---|
| Trailing adjusted EPS | $3.05 ÷ $1.72 | 1.77x |
| Trailing GAAP EPS | $0.757 ÷ $1.72 | 0.44x — the payout is 226% of GAAP earnings |
| Trailing free cash flow | ~$10,986M ÷ ~$9,802M | ~1.12x |
The GAAP figure is the least meaningful and the free-cash-flow figure is the most binding. dividendPayoutRatioTTM of 2.257 is arithmetically correct and describes the accounting, not the cash. What matters is that free cash flow exceeds the dividend by roughly 12% — the thinnest margin of any dividend-paying name in this batch — while adjusted earnings exceed it by 77%.
And the forward path is the reason this is Tactical rather than Core. At the 2029 consensus EPS of $2.304, the payout ratio on adjusted earnings is 74.7% and cover falls to 1.34x. That is survivable and it is not comfortable, and it assumes free cash flow tracks adjusted earnings rather than falling faster.
Total shareholder return from the dividend alone is 6.77% a year — no buyback is visible in the file for the trailing period — which means the base case's 10.2% price return and the yield together are roughly 17%, and roughly 40% of that return arrives in cash regardless of what the multiple does. For a 0.28-beta security, that is the entire investment argument in one sentence.
5. Valuation — priced in or room?
At $25.41, market capitalisation $144.82B, corrected enterprise value approximately $196.35B:
| Trailing (TTM to 2026-06-28) | 2026E | 2027E | 2028E | |
|---|---|---|---|---|
| Revenue | $63,696M | $61,774.8M (18 analysts) | $59,293.6M (18) | $54,737.8M (21) |
| Adjusted EPS | $3.05 | $2.943 (14) | $2.835 (13) | $2.438 (6 — thin) |
| GAAP diluted EPS | $0.757 | — | — | — |
| P/E on adjusted EPS | 8.33x | 8.63x | 8.96x | 10.42x |
Vendor priceToEarningsRatioTTM (GAAP) | 33.58x — REJECTED | — | — | — |
| EV / adjusted EBITDA | ~9.4x | — | — | — |
| Dividend yield | 6.77% | — | — | — |
| Free cash flow yield | 7.59% | — | — | — |
| Price / book | 1.70x | — | — | — |
| R&D as % of revenue | 20.1% | — | — | — |
Peer context. The vendor peer set is genuinely apposite: Merck ($316.2B), Amgen ($210.5B), Novo Nordisk ($196.8B), Gilead ($167.9B), Danaher ($136.9B), Bristol-Myers Squibb ($134.6B), Stryker ($128.8B), GSK ($103.2B), Sanofi ($102.5B), Boston Scientific ($73.0B). No peer multiples are supplied in the file, so no peer-multiple comparison is drawn. Our knowledge base supplies the one relevant relative observation, and it is from a source we have had to exclude for attribution reasons and therefore report at zero weight: a 2023-06-03 claim at conviction 45 that "on valuation Pfizer or Merck look preferable to Lilly — lower P/E (Lilly ~70x), higher dividend yield, and higher net margin/ROA."
5a. What today's price assumes (the inversion)
At $25.41 — 8.63x the 2026 consensus and 8.96x 2027 — the price embeds the following falsifiable claims, each with a number and a date:
- Adjusted EPS of $2.943 in 2026 and $2.835 in 2027. (Consensus-derived; 14 and 13 analysts.) The first half delivered $1.52, so the second half needs $1.42. The nearest test is 2026-11-03 against $0.817.
- Revenue declines 4.0% in 2027 and 7.7% in 2028. (Consensus-derived; 18 and 21 analysts.) The price does not assume growth. It assumes a managed decline, and the second-quarter print of +2.6% is ahead of that path.
- The dividend of $1.72 is maintained through the trough. (Our assumption; the trailing figure is filing-consistent and no forward rate is disclosed.) This is the most fragile assumption in the price. At the 2029 consensus it is 74.7% of adjusted earnings, and trailing free cash flow covers it by only about 12%. A cut would remove the entire reason to own this security and would almost certainly de-rate the multiple at the same time — the two risks are correlated, which is what makes it fragile rather than merely uncertain.
- "Other (income)/deductions—net" normalises. (Filing-disclosed at $3,716M in the quarter against $739M; our assumption on the direction.) If it does not, the GAAP series stays unusable and the equity keeps trading on a number many institutional screens will not accept.
- The market keeps paying 8-11x forward adjusted earnings. (Our number.) At 7x the 2027 consensus the stock is $19.85; at 12x it is $34.02. Pfizer traded above 15x forward as recently as its pre-2022 history, so the current 9x IS the cliff discount.
5b. The return bridge (why the multiple moves)
Expected return over the next twelve months decomposes as: adjusted EPS growth (MINUS 3.7%, from 2026E $2.943 to 2027E $2.835) + multiple drift (EXPANSION, from 8.63x on the current forward year to roughly 9.9x, +14%) + dividend yield (+6.77%) ≈ +17%.
Our base of $28 therefore requires the multiple to expand, and we name that as the fragile leg rather than burying it. $28 is 9.9x the 2027 consensus against today's 8.96x — a 10% re-rating on a declining earnings base.
The justification, and it is specific rather than generic. Earnings are falling on the consensus path, so a constant multiple produces a falling price — arithmetic that makes the yield the only source of return. The base assumes the multiple expands modestly because (i) the second-quarter beat of 12.9% on adjusted EPS suggests the consensus decline path is conservative; (ii) the dividend at 6.77% sets a valuation floor for as long as it is paid, and a 6.77% yield on a 0.28-beta security is a scarce instrument; and (iii) our knowledge base records generalist rotation into cheap pharmaceutical names as a live flow.
The honest alternative framing: at a FLAT 8.96x multiple on the 2027 consensus, the price in twelve months is $25.41 — exactly today — and the total return is the 6.77% dividend. That is the realistic downside case for a holder, and it is not a bad one. The bear case at $19 requires both a de-rating to 6.7x and the cliff proving deeper than modelled.
5c. Variant perception (where we differ, what would surprise)
- We differ from every screen on the leverage, and the correction is the largest dollar error in this batch's balance sheets. The published
netDebtof $66,274M andnetDebtToEBITDATTMof 4.93x ignore $12,454M of short-term investments and use an EBITDA denominator crushed by non-recurring charges. Corrected funded net debt is $51,529M at roughly 2.6x adjusted EBITDA, and enterprise value to adjusted EBITDA is roughly 9.4x rather than 16.4x. A screen ranking Pfizer on leverage is ranking it as one of the most indebted large-cap pharmaceuticals in the market. It is not. - We differ from the headline on where the revenue is coming from. Total revenue grew 2.6%. Product revenues FELL 0.8%; alliance revenues grew 18.7% and royalties 11.3%. Partnered and royalty income is now 21.1% of the total against 18.4% a year ago. That is invisible in the vendor payload —
seg_prodandseg_geoare both empty — and it changes how you read the cliff: Pfizer's owned portfolio is already flat, and the growth it does have is bought. Watchable number: product revenues in the 2026-11-03 release. - We differ from the GAAP reader entirely. The published trailing P/E of 33.58x is computed on GAAP earnings of $0.757 a share that include the $3,716 million "other deductions" line and its predecessors. On adjusted earnings the trailing multiple is 8.33x. The gap between 33.6x and 8.3x is the single widest GAAP-to-adjusted divergence in this batch, and it is why this security screens as expensive and trades as cheap.
- Where we do NOT have a variant perception: on the pipeline. No source in the data contract contains any pipeline information. We cannot assess whether Pfizer's 20.1% research intensity, its $10B+ of business-development capacity, its $1.25 billion licence of 3SBio's SSGJ-707 or its contested acquisition of Metsera will replace the revenue the cliff removes. We are buying a yield and a discount, not a scientific view, and the dive says so.
- Positive surprise that would force a re-rate: product revenues returning to growth on 2026-11-03; a disclosed normalisation of the "other deductions" line, which would restore a usable GAAP series and bring the screen-driven buyers back; or a dividend increase, which would settle the cover question in the strongest possible way.
- Negative surprise that would break the thesis: any signal on the dividend — a hold at $1.72 without an increase is neutral, a cut is thesis-ending; free cash flow falling below the roughly $9.8 billion the payout costs; a goodwill or intangible impairment against the $124,995 million carried; a large debt-funded acquisition into a declining earnings base; or the chief financial officer transition on 2026-08-15 being followed by guidance revision.
Synthos fair values
All three anchors are multiples of the 2027 consensus adjusted EPS distribution (mean $2.835, low $2.606, high $3.004, 13 analysts) — the last forward year with double-digit EPS coverage.
- Bear ~$19 — 6.7x the 2027 consensus mean, cross-checked at 7.3x the consensus LOW of $2.606 and 7.8x the 2028 mean; roughly 19% below the 52-week low of $23.53. The scenario: the cliff proves deeper than the 2028 consensus of $2.438, free cash flow falls below the dividend, the payout is cut, and the multiple compresses to the level a declining, non-yielding pharmaceutical commands. −25.2%.
- Base ~$28 — 9.9x the 2027 consensus MEAN of $2.835. Cross-check: 9.5x the 2026 consensus; 11.5x the 2028 consensus. Sensitivity, stated openly: 9x the 2027 consensus gives $25.52 — essentially spot — and 11x gives $31.19. The answer lives in a 9-11x band. The scenario: the decline path runs to consensus or slightly better, the dividend is maintained, "other deductions" normalises, and the multiple expands modestly as the yield attracts capital. +10.2%, plus a 6.77% dividend.
- Bull ~$35 — 12.3x the 2027 consensus mean, cross-checked at 11.7x the consensus HIGH of $3.004 and 14.4x the 2028 mean; roughly 23% above the 52-week high of $28.55. The scenario: business development replaces the cliff faster than modelled, product revenues return to growth, the 2028 consensus of $2.438 proves the floor rather than the path, and the market re-rates a 6%-yielding pharmaceutical toward a market-average multiple. +37.7%.
Base is 10.2% above spot; asymmetry approximately 1.50:1 to the upside (25.2% down, 37.7% up). Add the 6.77% dividend and the base-case total return is roughly 17% on a 0.28-beta security — which is the argument, and it is an income argument rather than a growth one.
6. Knowledge base — 28 claims, and the attribution defect is confirmed here
Raw hits: 28 (entity matches on Pfizer). Channels: 6. Excluded for attribution: 6. Excluded as management voice: 2. Concentration test failed. Effectively usable: the biotech_hangout block plus three peripheral claims.
The cross-company attribution failure, confirmed
SIX of the twenty-eight claims come from the channel slug eli_lilly_ceo and every one of them is about PFIZER:
> 2020-12-08 · bullish · conviction 85 · channel: eli_lilly_ceo — "Pfizer/BioNTech vaccine efficacy is stellar; ~$2B invested (self-funded), targeting 50M doses in 2020 and 1.3B in 2021; already deployed in UK."
> 2020-09-04 · bullish · conviction 75 — "Pfizer-BioNTech mRNA vaccine pivotal trial (30,000 enrollees) could read out efficacy by end-October with rolling FDA submission."
> 2020-09-04 · bullish · conviction 75 — "Remdesivir supply scaled ~40-fold... via 30+ partners including Pfizer."
> 2021-06-19 · neutral · conviction 72 — on intellectual-property waivers and manufacturing know-how.
> 2023-06-03 · bearish · conviction 45 — "On valuation Pfizer or Merck look preferable to Lilly — lower P/E (Lilly ~70x), higher dividend yield."
> 2011-12-17 · bearish · conviction 75 — on the 2011-2013 patent cliff.
The 2020-12-08 claim is decisive. "~$2B invested (self-funded)" is a statement about Pfizer's own internal financing decision on its own vaccine programme — the kind of detail a company's management discloses about itself, not the kind a competitor's chief executive volunteers. Under a channel slug named for the chief executive of Eli Lilly.
This is the documented systemic defect the attribution policy exists to catch: one channel slug named for a pharmaceutical chief executive carrying other companies' management voice. And it is confirmed independently in this same batch: the identical eli_lilly_ceo slug appears in the Uber lane carrying two claims — that Uber's exit from India "cleans up its P&L" and that COVID would permanently shift Americans to online ordering. A channel named for a pharmaceutical executive is carrying claims about a ride-hailing company.
> All six eli_lilly_ceo claims are EXCLUDED from the conviction pool, regardless of content. They are quoted above only to document the failure.
The management-voice claims with no tag
Two of the three lex_fridman claims, both dated 2021-12-18, are unmistakably Albert Bourla — Pfizer's chairman and chief executive — speaking about Pfizer:
> conviction 80 — "Under urgency, run drug-development phases in parallel and spend to compress timelines (risked $2.3B up front), trading capital efficiency for speed."
> conviction 75 — "Focus on the real value driver—breakthroughs that change patients' lives—and financial returns follow; chasing money directly fails."
No speaker_role is set. This is the third instance in this batch of the management-voice flag failing to fire — after Glenn Fogel on Booking Holdings and the Andrew McDonald block on Uber — while it fired correctly for Thomas Peterffy on Interactive Brokers. Both claims are excluded from the conviction pool and half-weighted.
The third lex_fridman claim IS a genuine third-party bear point and is used:
> 2022-02-11 · bearish · conviction 80 — "Pfizer found guilty of fraud and racketeering in a 2010 federal trial; drug companies routinely violate marketing and data-reporting laws."
Four years old, historical rather than forward-looking, and recorded because reputational and legal tail risk is a genuine sector feature.
The concentration test, which the lane fails
biotech_hangout supplies 16 of 28 — 57.1%. Stripping it, and applying the two exclusions above, leaves: one genuine third-party bear claim (lex_fridman on the 2010 conviction), one China-competition claim (all_in), one academic-funding claim (natalie_brunell), and one mild biotech-bullish claim (jacob_shapiro). The lane does not survive its concentration in any meaningful form.
What is left, and it is the useful part
The biotech_hangout block is sector-expert commentary rather than security analysis, and it contains the two most specific facts in the lane — both bearish:
> 2025-08-28 · bearish · conviction 60 · horizon: FACT — "Pfizer quantified Part D redesign as ~1.6% top-line hit ($1.5B drag, $500M boost); most other pharmas won't disclose similar granularity."
> 2025-08-27 · bearish · conviction 70 — "IRA Part D redesign raises payer liability on catastrophic coverage from 15% to 60%, driving outright payer nos on expensive specialty drugs; a broad risk across the sector."
The first is the most useful claim in the entire lane: a company-quantified headwind of roughly 1.6% of revenue, with the components separated. The second is the mechanism, and it is a sector-wide claim rather than a Pfizer-specific one.
On business development, which is the answer to the cliff:
> 2025-08-26 · bullish · conviction 70 — "Pfizer BD capacity rising (10-15 deals, higher next year), focused on internal medicine and I&I rather than vaccines or large oncology."
> 2025-08-27 · bullish · conviction 75 — "2025 shaping up as strong M&A year; Pfizer has $10B+ to deploy, deals shifting from small add-ons to larger acquisitions."
> 2025-08-21 · bullish · conviction 60 · FACT — "3SBio's SSGJ-707 ASCO data ($1.25B upfront to Pfizer) looked genuinely impressive; ~67.5% ORR, near-100% disease control, deepening responses—among best in class."
On Metsera, and note that the three claims record a live bidding war rather than a settled fact:
> 2025-10-31 · neutral · conviction 60 — "Novo's ~$9B counterbid tops Pfizer's $7.3B for Metsera (~19% better/share) but faces serious FTC/antitrust overlap risk."
> 2025-11-14 · neutral · conviction 60 — "Pfizer aggressively won Metsera to find post-pandemic footing."
> 2025-11-14 · neutral · conviction 45 — "FTC threatened to sue over the Metsera deal structure."
The claims are internally inconsistent on the outcome — one says Novo's bid "tops" Pfizer's, a fortnight later another says Pfizer "won" — which is what a contested transaction looks like in a knowledge base built from dated sources. We record the sequence rather than picking a winner, and note that no filing in this archive confirms the outcome.
And the most recent relevant claim, which describes the current setup rather than the company:
> 2026-01-16 · bullish · conviction 68 — "Generalists are rotating into pharma/biotech because it looks cheap versus an expensive AI-driven S&P; interest is broadening beyond mega-caps to Merck, Bristol, Pfizer, Gilead, Amgen."
A flows argument, not a fundamental one, and treated as such. Against it: all_in (2026-01-15, bearish, conviction 80) — "US is losing the biotech race to China (5%→42% of licensing deals since 2022); has ~3-year window to reform or fall behind on speed" — a claim Pfizer's own $1.25 billion licence of a Chinese asset both illustrates and, from Pfizer's specific point of view, mitigates.
Conclusion. The Synthos knowledge base has no reliable name-level view on Pfizer as a security. Its most specific content is sector commentary from one channel; six claims are excluded for a confirmed cross-company attribution failure; two are unlabelled management voice. Conviction rating: Low. Breadth: 6 nominal, effectively 2. Net conviction: mildly positive on flows, mildly negative on policy, and the lane collapses under its own tests. No conviction premium is applied.
7. Data integrity — what we rejected from the vendor file and why
Nine findings.
1. netDebt ignores $12,454M of short-term investments — REJECTED and rebuilt. The vendor's $66,274M is total debt of $67,416M less cash of $1,142M only. The correct figure, counting the short-term investments the balance sheet carries as liquid, is $53,820M, or $51,529M excluding $2,291M of lease obligations. This is the standing omission from the defect register and, at $14.7 billion, the largest dollar instance in this batch.
2. netDebtToEBITDATTM of 4.93x and evToEBITDATTM of 16.40x are struck on a charge-depressed EBITDA — RECOMPUTED. The vendor's implied trailing EBITDA of roughly $13,447M reflects a quarterly series of $5,648M, $756M, $5,021M and $1,199M, with two of four quarters crushed by "Other (income)/deductions—net". On trailing operating income of $15,976M plus roughly $4,900M of depreciation and amortisation — adjusted EBITDA of about $20,876M — corrected leverage is roughly 2.6x and EV/EBITDA roughly 9.4x. Method stated; no greater precision claimed.
3. priceToEarningsRatioTTM of 33.58x is computed on unusable GAAP earnings — REJECTED. Trailing GAAP EPS of $0.757 includes a June-quarter "Other (income)/deductions—net" charge of $3,716M against $739M a year earlier, producing a $0.04 GAAP LOSS against $0.77 of adjusted EPS. On adjusted earnings the trailing multiple is 8.33x. The 25-point gap between the two is the widest GAAP-to-adjusted divergence in this batch. Also rejected: earningsYieldTTM, dividendPayoutRatioTTM (2.257) and effectiveTaxRateTTM (−3.90%).
4. seg_prod and seg_geo are BOTH EMPTY, across a segment reorganisation. Pfizer discloses revenue by product portfolio and by geography, and neither is in the payload. Worse, the 8-K of 2026-08-04 records that "In the first quarter of 2026, we made changes in our commercial structure, which included the transition of certain off-patent branded and generic sterile injectables and biosimilars primarily from the Specialty Care and Oncology product portfolios to a new Hospital and Biosimilars product portfolio within our Biopharma reportable segment. Effective January 1, 2026..." A reader working from the payload alone cannot see the portfolio structure, let alone that it changed.
5. est.ebitdaAvg and est.ebitAvg show the fixed-ratio signature — NOT USED. From 2026 through 2030, ebitdaAvg is 29.49% of revenueAvg and ebitAvg 21.32% in every year. Independently, the 2025 ebitAvg of $13,202.7M is 14.5% below actual operating income of $15,437M. All forward valuation runs on epsAvg and revenueAvg.
6. The 2023 estimate row is wrong by more than a sign. It reports netIncomeAvg of MINUS $2,593M and epsAvg of MINUS $0.459 against actual 2023 net income of $2,133M and EPS of $0.38. Not used, and recorded as a reminder of how wide the historical error bars on this business have been.
7. epsdiluted is NULL on every row and eps is the basic figure. For the June 2026 quarter the filing reports basic and diluted EPS both at ($0.04); for the half, both at $0.43. All EPS figures here are filing-sourced or clearly labelled as consensus.
8. lastDividend is absent and no filing in this archive states the declared quarterly rate. profile.lastDiv is null and dividendPerShareTTM of $1.72 is the trailing-twelve-month total. We assert no forward run-rate and use the trailing figure throughout, labelled as trailing. On a security whose entire investment case is the dividend, the absence of a declared-rate disclosure in the archive is a material gap and is stated as one.
9. quote.yearHigh/yearLow ($28.75 / $23.58) disagree with tech.hi52/lo52 ($28.55 / $23.53) — we use tech. The discrepancies are 0.7% and 0.2%. Note separately that tech.max_dd_from_peak reads −58.5% against pct_from_hi of −11.0%: the six-year peak sits far outside the trailing year, and both figures are correct on their own basis. This is NOT the defect the register warns about — it is a genuine reading, and it tells you this stock is 58.5% below its 2021-era high.
Not defects, correctly reported: the share count reconciles (market capitalisation of $144,822,770,400 ÷ $25.41 = 5,699.4M against the filing's 5,699M weighted-average diluted); revenuePerShareTTM reconciles to filing-derived trailing revenue; researchAndDevelopementToRevenueTTM of 20.08% is accurate and is the highest in this batch; interestCoverageRatioTTM of 5.92x is consistent; priceToBookRatioTTM of 1.70x is meaningful here, unusually, because Pfizer's equity is real rather than buyback-depleted; and earn_cal for 2026-08-04 correctly carries the actual — $0.77 against $0.682 estimated and $15,034M against $14,399M — on the same day the 10-Q was filed, which is the fastest refresh observed in this batch.
Non-equity tripwire — checked and passed. PFE is common stock, NYSE-listed, beta 0.28 — the lowest in this batch and low enough to warrant the check — but the security is confirmed as common equity by a variable dividend (the trailing $1.72 is not a fixed coupon), a 52-week range of $23.53 to $28.55 (21.3%), a six-year drawdown of 58.5%, volume of 55.4 million shares (roughly $1.41B of turnover) and 5,699 million shares outstanding. A 0.28 beta on a $145 billion company is a defensive equity, not a fixed-income instrument.
8. Technicals
- Price $25.41. −11.0% from the 52-week high of $28.55; +8.0% above the 52-week low of $23.53. Position within the annual range: 37th percentile. The six-year maximum drawdown from peak is −58.5%, so the stock remains less than half its 2021-era value.
- The moving averages straddle the price: +1.4% above the 50-day at $25.06, −1.6% below the 200-day at $25.84. A flat, range-bound chart — the least directional in this batch.
- RSI 59.1 — mildly constructive. MACD +0.08 — essentially zero.
- Relative performance is the least bad of the four laggards here: 3-month −3.4% vs SPY +7.6%; 6-month −4.7% vs SPY +11.1%; 12-month +8.2% vs SPY +24.3% and QQQ +30.8%. A 16-point twelve-month shortfall against the index — meaningful, but the only positive absolute twelve-month return among the batch's underperformers.
- Sentiment: 15 buy, 23 hold, 1 sell across 39 analysts, consensus "Hold". Target $26.88 (+5.8%), median $27.00 (+6.3%), high $30 (+18.1%), low $25 — 1.6% BELOW spot. A 5.8% consensus-implied upside is the second-smallest in this batch after Blackstone's, and a Hold consensus with a low target below spot says the sell side sees the yield and not much else.
Today's move
PFE closed 2026-08-04 at $25.41, up $0.38 or 1.52% from $25.03, on 55,415,165 shares — roughly $1.41 billion of turnover and the second-heaviest volume in this batch. It opened at $25.04, traded $24.70 to $25.64, and closed near the upper end.
And unlike Booking Holdings in this same batch, the market clearly saw the numbers. The company reported second-quarter results on 2026-08-04 and the earnings calendar has already captured the actuals — adjusted EPS $0.77 against $0.682 and revenue $15,034M against $14,399M. A 1.52% gain on double-normal volume against a 12.9% earnings beat and a GAAP loss is a measured reaction: the market took the adjusted number and discounted the GAAP one, which is the correct reading.
9. Insiders — accruals only, and no transaction of substance
Eight records, all A-Award, all phantom stock units:
| Person | Role | Security | Units | Date | Holding after |
|---|---|---|---|---|---|
| Albert Bourla | Chairman & CEO | Phantom Stock Units SSP | 24 | 2026-07-31 | 786,154 |
| Albert Bourla | Chairman & CEO | Phantom Stock Units SSP | 24 | 2026-07-15 | 772,901 |
| Albert Bourla | Chairman & CEO | Phantom Stock Units SSP | 84 | 2026-07-09 | 772,938 |
| Albert Bourla | Chairman & CEO | Phantom Stock Units SSP | 25 | 2026-06-30 | 772,979 |
| Shantanu Narayen | Director | Phantom Stock Units | 2,058.46 | 2026-06-26 | 200,879 |
| James C. Smith | Director | Phantom Stock Units | 1,852.61 | 2026-06-26 | 182,886 |
| Joseph Echevarria | Director | Phantom Stock Units | 1,904.08 | 2026-06-26 | 165,632 |
| Mortimer J. Buckley | Director | Phantom Stock Units | 1,543.85 | 2026-06-26 | 26,335 |
The reading is short. Every entry is a zero-price award of phantom stock units. The chief executive's four entries — 24, 24, 84 and 25 units at semi-monthly intervals — are dividend-equivalent accruals on a savings-plan balance, not purchases. The four directors received simultaneous quarterly phantom-stock grants on 2026-06-26.
There is not a single open-market purchase or open-market sale in this file, and no F-InKind withholding either. That is a neutral signal and it is reported as neutral.
The one item worth recording is the size of the chief executive's holding: 786,154 phantom stock units, worth roughly $20.0 million at the closing price, accumulating steadily. And the one item worth watching is not in this table at all: the chief financial officer, Dave Denton, departs on 2026-08-15.
10. Verdict, kill-criteria and flip conditions
Buy — Tactical.
What is genuinely attractive, and most of it is arithmetic rather than opinion: a 6.77% dividend yield, the highest in this batch by nearly three percentage points, on a security with a beta of 0.28, the lowest here by more than half; 8.33x trailing adjusted earnings and 8.63x the 2026 consensus; a 7.59% free-cash-flow yield; a second-quarter print filed today that beat by 12.9% on adjusted EPS and 4.4% on revenue — the fifth consecutive beat; revenue up 2.6% in the quarter and 3.9% for the half against a consensus that models decline; alliance revenues up 18.7%; research and development at 20.1% of revenue, the highest reinvestment rate in this batch; and a corrected balance sheet at roughly 2.6x adjusted EBITDA against the 4.93x every screen shows.
What keeps this Tactical rather than Core: consensus itself models EPS falling from $2.943 to $2.438 by 2028 and revenue falling 13.5% by 2029; free cash flow covers the dividend by only about 1.12 times, the thinnest margin in this batch, and the dividend is the investment case; GAAP results are unusable, with a $3,716 million "other deductions" line the filing does not disaggregate turning $4.7 billion of operating income into a loss; tangible equity is roughly negative $38.5 billion against $124,995 million of goodwill and intangibles on assets facing expiry; the chief financial officer departs in eleven days; the sell side is a Hold with a low target below spot; our base case requires a 10% multiple expansion on a declining earnings base; and our knowledge base collapses under its own tests, with six claims excluded for a confirmed cross-company attribution failure and two more for unlabelled management voice.
The distinction that matters. This is not a bet that Pfizer grows. Consensus says it shrinks, we agree, and the price says so too at 8.6x. It is a bet that a 6.77% dividend on a 0.28-beta security is maintained through a decline that is already forecast, and that the market pays slightly more than 9x for that once it stops falling. Roughly 40% of the base-case return arrives in cash regardless of what the multiple does, which is a materially different risk profile from anything else in this batch. The whole thesis reduces to one number — free cash flow against the dividend — and that number currently has a 12% margin. Size accordingly.
Pre-registered KILL criteria — what would take this to Avoid:
- Any reduction in the dividend, or trailing free cash flow falling below roughly $9.8 billion. This is the thesis and it is a single, published, quarterly-observable number.
- Adjusted EPS guidance or consensus for 2028 falling below roughly $2.20, at which the payout ratio exceeds 78% and cover falls under 1.3x.
- A goodwill or intangible impairment against the $124,995 million carried, which on assets facing patent expiry is the specific risk the negative tangible equity represents.
- A large debt-funded acquisition taking corrected funded net debt above roughly 3.5x adjusted EBITDA into a declining earnings base.
- Product revenues declining more than 3% in a quarter, from −0.8%, which would mean the owned portfolio is eroding faster than the partnered one can offset.
- A second consecutive quarter of "Other (income)/deductions—net" above $3 billion without disaggregation.
Pre-registered FLIP conditions — what would take this to Buy — Core:
- A dividend increase. This is worth a full tier on its own, because it would settle the cover question in the only way management can — and because the archive currently contains no declared-rate disclosure at all.
- Product revenues returning to growth on 2026-11-03, which would say the cliff is being outrun rather than merely funded.
- Disaggregation of the "Other (income)/deductions—net" line showing it to be non-recurring, which would restore a usable GAAP series and bring back the large pool of buyers whose screens will not accept a 33.6x trailing multiple.
- A price below approximately $22 (roughly 7.8x the 2027 consensus, below the 52-week low), which would put the yield above 7.8% and the base case at +27%.
- A genuine, correctly attributed, name-level knowledge-base claim. At present six of twenty-eight claims arrive under a competitor's chief executive's channel slug and two more are unlabelled management voice; the lane is not currently usable as evidence.
Where PFE fits in the Synthos Framework Portfolio. The defensive-income sleeve alongside AT&T, at 0% today with a 2-3% target initiated here. On batch overlap, and it is a real point: Pfizer and AT&T are the two income positions in this batch, yielding 6.77% and 4.75% on betas of 0.28 and 0.42 — and they are not the same trade. AT&T's cash flow is contracted and its risk is a $23 billion spectrum purchase; Pfizer's cash flow is patent-protected and its risk is that the patents expire on a published schedule. AT&T's dividend is covered 2.5 times by free cash flow; Pfizer's by 1.12 times. If only one income position is taken, AT&T has the safer payout and Pfizer the higher yield and the cheaper multiple, and the choice is a straightforward risk preference rather than a research question. Logged as a tracked Synthos call (Buy — Tactical) as of 2026-08-04 at $25.41.
Single biggest risk: dividend cover. The payout costs approximately $9.8 billion a year against trailing free cash flow of approximately $11.0 billion — a margin of about 12%, the thinnest in this batch. On adjusted earnings the cover is a more comfortable 1.77 times, and even at the 2029 consensus EPS of $2.304 it would be 1.34 times. But those are earnings, and dividends are paid in cash. A 6.77% yield on a 0.28-beta security is the entire reason to own this equity; if it is cut, the multiple and the yield reprice together, which is why the bear case at $19 is a 25% decline rather than a 7% one. Everything else in this dive — the cheap multiple, the corrected leverage, the research intensity, the beat — is secondary to whether free cash flow stays above $9.8 billion.
Provenance & disclosures
- Traceability: 28 tagged knowledge-base claims name Pfizer across 51,928 distilled claims, from 6 channels. THE DOCUMENTED CROSS-COMPANY ATTRIBUTION FAILURE IS CONFIRMED HERE AND IS THE MOST IMPORTANT FINDING IN THIS SECTION: SIX claims arrive under the channel slug
eli_lilly_ceoand ALL SIX are about PFIZER, including a 2020-12-08 claim at conviction 85 stating the Pfizer/BioNTech vaccine was "~$2B invested (self-funded)" — Pfizer management framing about Pfizer's own financing, carried under a slug named for a competitor's chief executive. The SAME slug appears in the Uber lane in this batch carrying two claims about Uber's India exit and pandemic delivery habits, i.e. a channel named for a pharmaceutical executive carrying claims about a ride-hailing company. That is two independent confirmations, and ALL SIX Pfizer claims are EXCLUDED from the conviction pool regardless of content. A SECOND exclusion: two of the threelex_fridmanclaims (2021-12-18, convictions 80 and 75 — "risked $2.3B up front", "focus on the real value driver") are unmistakably Albert Bourla, Pfizer's chairman and chief executive, speaking about Pfizer, with NOspeaker_roletag set. This is the THIRD instance in this batch of the management-voice flag failing to fire (after Glenn Fogel on Booking Holdings and the nine-claim Andrew McDonald block on Uber), while it fired correctly for Thomas Peterffy on Interactive Brokers. Both are excluded from the pool and half-weighted. Concentration sensitivity test run and reported:biotech_hangoutsupplies 16 of 28 (57.1%); stripping it, after the exclusions above, leaves FOUR peripheral claims — one genuine third-party bear point (lex_fridman, 2022-02-11, conviction 80, on the 2010 federal fraud and racketeering conviction), one China-competition claim (all_in, 2026-01-15, conviction 80), one academic-funding claim (natalie_brunell) and one mild biotech claim (jacob_shapiro). The lane does NOT survive its concentration. What remains substantive is entirelybiotech_hangoutsector commentary, whose two most useful items are BEARISH and specific: Pfizer's own quantification of the Inflation Reduction Act Part D redesign at roughly 1.6% of the top line ($1.5B drag, $500M boost), and the payer-liability shift on catastrophic coverage from 15% to 60%. Three claims on the Metsera acquisition are internally inconsistent on the outcome (one says Novo's bid "tops" Pfizer's, a fortnight later another says Pfizer "won") and are recorded as a sequence rather than resolved, since no filing in this archive confirms it. All quotes are verbatim from the stored claim text. - Data as-of: fundamentals — second-quarter 2026 income statement as at 2026-06-28, from the 10-Q filed 2026-08-04, THE DATE OF THIS DIVE; balance sheet as at 2025-12-31 from the payload's annual block, cross-referenced to the 10-K filed 2026-02-26 · estimates 2026-08-04 · prices 2026-08-04, quote timestamp 1785873758 = 2026-08-04T20:02:38Z ($25.41, +1.52%; 50-DMA $25.06; 200-DMA $25.84; RSI 59.1; MACD +0.08) · knowledge-base claims 2026-08-04. Fiscal-year note: Pfizer's fiscal year is the calendar year but its quarters end on the Sunday nearest the calendar quarter-end — the "second quarter of 2026" ended 2026-06-28 and the prior-year comparative 2025-06-29. All figures come from the Synthos vendor data file for PFE or from the SEC filings in the PFE archive; no figure comes from memory, recall or external retrieval.
- Filing archive contents: 10-K filed 2026-02-26 (fiscal 2025, 765,701 characters, financial-statement tables preserved); 10-Q filed 2026-05-05 (March 2026 quarter); 10-Q filed 2026-08-04 (June 2026 quarter, 287,918 characters — THE PRIMARY SOURCE, filed on the dive date); two 8-Ks filed 2026-06-18, the substantive one being Item 5.02 — chief financial officer Dave Denton's notice on 2026-06-16 that he steps down effective 2026-08-15 for a role "outside of the pharmaceutical industry in consumer goods", with the express statement that the departure "is not related to the Company's financial or operating results or to any disagreements"; 8-K filed 2026-08-04 (announcing second-quarter results, and separately documenting the first-quarter-2026 commercial reorganisation creating a new "Hospital and Biosimilars" product portfolio within the Biopharma reportable segment).
- Where the filings contradicted or corrected the vendor (detailed in Section 7):
netDebtrebuilt from $66,274M to $51,529M funded after identifying the omission of $12,454M of short-term investments — at $14.7 billion the largest dollar instance of this defect in the batch;netDebtToEBITDATTMrecomputed from 4.93x to roughly 2.6x andevToEBITDATTMfrom 16.40x to roughly 9.4x on a constructed adjusted EBITDA of about $20,876M, method disclosed;priceToEarningsRatioTTMof 33.58x REJECTED in favour of 8.33x on adjusted earnings, because trailing GAAP EPS of $0.757 includes a June-quarter "Other (income)/deductions—net" charge of $3,716M against $739M that produced a $0.04 GAAP LOSS against $0.77 of adjusted EPS;seg_prodandseg_georecorded as BOTH EMPTY across a first-quarter-2026 commercial reorganisation disclosed in the 8-K;est.ebitdaAvg/ebitAvgrejected on fixed revenue ratios of 29.49% and 21.32% across five years and a 2025 EBIT estimate 14.5% below actual operating income; the 2023 estimate row identified as wrong by more than a sign (consensus net income of MINUS $2,593M against an actual PLUS $2,133M);epsdilutedNULL on every row;dividendPerShareTTMof $1.72 identified as the trailing total with NO declared rate available anywhere in the archive, so no forward run-rate is asserted — a material gap on a security whose case is the dividend; andquote.yearHigh/yearLowrejected in favour oftech. Where vendor and filing AGREED — worth recording: the share count (5,699.4M implied against the filing's 5,699M weighted-average diluted);revenuePerShareTTM;researchAndDevelopementToRevenueTTMof 20.08%;interestCoverageRatioTTM; and — notably —earn_calcarried the actual for 2026-08-04 on the same day the 10-Q was filed, $0.77 against a $0.682 estimate and $15,034M against $14,399M, the fastest data refresh observed anywhere in this batch. - Basis note: Pfizer reports GAAP results and an adjusted diluted EPS measure. Analyst estimates and the earnings-calendar actuals are struck on ADJUSTED EPS. The GAAP-to-adjusted divergence here is the widest in this batch: the June quarter reported MINUS $0.04 GAAP against PLUS $0.77 adjusted, an $0.81 gap equal to 105% of the adjusted figure, entirely attributable to a "Other (income)/deductions—net" line of $3,716M that the extracted filing does not disaggregate. All forward multiples use consensus adjusted EPS; the trailing multiple is shown on the adjusted basis at 8.33x, and the vendor's GAAP-based 33.58x is rejected rather than presented alongside, because on this security it is an artefact that would mislead in the opposite direction from every other name in this batch.
- Estimate coverage: 18 analysts on 2026 revenue and 14 on EPS; 18 and 13 on 2027; 21 and SIX on 2028; 10 and FOUR on 2029; 10 and TWO on 2030. The fair value is struck on the 2027 EPS distribution (13 analysts), the last forward year with double-digit coverage. The 2028 EPS row is used only as a directional cross-check and its dispersion is reported prominently — $1.578 low against $2.780 high on six analysts, a range equal to 49% of the mean, which is genuine disagreement about the depth of the cliff. The 2029 and 2030 EPS rows are excluded from every conclusion.
- Peer note: the vendor peer set is genuinely apposite — Merck ($316.2B), Amgen ($210.5B), Novo Nordisk ($196.8B), Gilead ($167.9B), Danaher ($136.9B), Bristol-Myers Squibb ($134.6B), Stryker ($128.8B), GSK ($103.2B), Sanofi ($102.5B), Boston Scientific ($73.0B). No peer multiples are supplied in the file, so no peer-multiple comparison is drawn. The only relative-valuation observation available in the knowledge base comes from an excluded source and is reported at zero weight.
- Fair-value caveat: the $19 / $28 / $35 anchors are multiples of the 2027 consensus adjusted EPS distribution — 6.7x the mean of $2.835 for the bear (cross-checked at 7.3x the low of $2.606), 9.9x the mean for the base, and 12.3x the mean for the bull (cross-checked at 11.7x the high of $3.004). Stated arithmetic, not a discounted cash flow. The base is sensitivity-disclosed: 9x gives $25.52, essentially spot, and 11x gives $31.19. THE BASE REQUIRES MODEST MULTIPLE EXPANSION and this is named as the fragile leg: $28 is 9.9x the 2027 consensus against today's 8.96x, a 10% re-rating on a DECLINING earnings base. The honest alternative framing is stated in Section 5b: at a flat 8.96x multiple the price in twelve months is today's price and the total return is the 6.77% dividend — which is the realistic holder's downside and is not a bad one.
- Timing: SECOND-QUARTER 2026 RESULTS WERE RELEASED ON 2026-08-04, THE DATE OF THIS DIVE, with the 10-Q filed the same day and the actuals already captured in the earnings calendar. Unlike Booking Holdings in this batch, the market clearly reacted: +1.52% on 55.4 million shares, roughly double normal volume. The next scheduled print is 2026-11-03, 91 days away, against $0.817 of adjusted EPS on $16,153.4M of revenue. A nearer, unscheduled event sits in between: the chief financial officer's departure on 2026-08-15, eleven days after this dive, disclosed on 2026-06-18 with the standard non-disagreement language.
- Accessibility note: no information in this dive is conveyed by colour. All emphasis is carried by bold text, table structure and explicit labelling.
- Not investment advice. Independent research, educational and informational only, never personalised. No recommendation to buy, sell or hold any security is made to any person.
- Version: 2026-08-04-full.