Alnylam Pharmaceuticals [ALNY] — Valuation

2026-07-29 · framework: Criteria, 2026-07-29 Two outputs are mandatory and both are produced. Reporting only one is the defect this structure prevents.

Part A — Inputs, every one verified against primary filings

Recency asserted. Latest 10-Q filed 2026-04-30 for the period ended 2026-03-31 — 90 days old. The FY2025 10-K was filed 2026-02-12. Neither is stale. (The XBRL companyfacts feed for this issuer is stale — max filed date 2026-04-06 — which is what produced the screen's 301-day-old revenue. The R-file route used here is current. See ALNY_Research.md §0.)

TTM used, never last-FY. Last-FY revenue ($3,713.9m) would understate the current base by 13.4% and would misstate the multiple by the same. The rule exists because last-FY understated MU by 136%.

Part B — The implied-path test (the Valuation Criteria)

reverse_dcf.py is a pure terminal-value instrument: terminal EV = revenue_T × terminal margin × exit multiple, discounted at WACC. Terminal value is therefore 100% of EV by construction, which is why the reverse DCF is the primary long-horizon output and the forward DCF is not run as the verdict.

B1 — The exit multiple, and its basis

Basis: GROWTH_MATCHED. An exit multiple may only be drawn from a comparator set whose growth brackets the subject's growth at the exit year. ALNY's projected FY2031 revenue growth is in the 12–18% range (decelerating from the guided +64–77% in FY2026 as the ATTR-CM diagnosed pool is worked through, with nucresiran launching by 2030). The comparator set below spans 1.0% to 21.4% and brackets it on both sides.

Computed 2026-07-29 from Alpaca prices and EDGAR FY2025 filings, one method, one date:

Exclusions, stated. JAZZ and SRPT have negative GAAP operating income — an EBIT multiple is undefined, not zero. REGN is excluded on a data defect, not judgement: its EntityCommonStockSharesOutstanding tag in EDGAR is dated 2012-07-13, giving 94.0m shares against a real count materially higher. Its 16.4x is computed on an understated market cap and is not trustworthy. This is the same staleness class as the ALNY revenue defect and is disclosed rather than silently averaged in.

Base exit multiple: 27.4x EBIT, the median of the growth-matched set.

No distant-year haircut is applied, deliberately. valuation.md: "The base exit multiple may not sit below every stated anchor without a separately argued reason." The NTRA failure was naming anchors of 7.0x and 7.5x and setting base at 5.0x — a haircut stacked on a multiple that had already mean-reverted. These anchors (27.3–28.2x) are current market multiples for large-cap biopharma with intact franchise exclusivity, which is ALNY's situation at the 2031 exit year (AMVUTTRA is a 2022-approved product; nucresiran launches by 2030). Setting base below all three would repeat that defect. The sensitivity below carries the compression case explicitly instead.

B2 — The terminal margin, and its basis

35% GAAP EBIT margin. Derived, not assumed:

* FY2026 guidance: total net product revenue $4,900–5,300m plus collaborations and royalties $400–500m = $5,300–5,800m (mid $5,550m); non-GAAP R&D + SG&A $2,700–2,800m plus $300–400m of excluded SBC = GAAP R&D + SG&A of roughly $3,100m; COGS at ~20% of product revenue = ~$1,020m. Implied GAAP operating income ≈ $1,415m on $5,550m.

35% assumes ALNY reaches Vertex's structural margin by 2031 — five years after its first profitable year, with revenue roughly doubling again and R&D held near 20% of sales. Sensitivity is run over it.

B3 — The result

python3 reverse_dcf.py --spot 288.50 --shares 138.226 --net-cash 324.9 \
    --revenue 3986.9 --years 5 --wacc 0.10 \
    --terminal-margin 0.35 --exit-multiple 27.4 --hist-cagr 0.5185

On the total TTM base ($4,286.9m, including the $300m Roche recognition) the requirement is 9.2% and the margin against the 52.97% total demonstrated CAGR is +43.8pp. The recurring run is the one that governs; the total run is shown so the choice is visible.

B4 — Implied compression, stated as a number

The price is not requiring multiple expansion. It requires the multiple to hold roughly flat while revenue compounds at 10.7% — against a business that just guided +49% for the coming year.

B5 — Sensitivity, run over the exit multiple (never over scenario probabilities)

Recurring base $3,986.9m, terminal margin 35%, WACC 10%, 5 years.

Second sensitivity, over the terminal margin (the other high-variance parameter), at 27.4x:

Worst corner tested: 15.0x exit AND 25% terminal margin → required CAGR 33.6%. Even there the requirement sits 18.3pp below the demonstrated recurring CAGR. There is no plausible corner of this parameter space in which today's price requires more than ALNY has already delivered.

B6 — What could break the PASS

The test passes on every parameterisation tried. The honest statement of its weakness is therefore not about the parameters but about the base: the required 10.7% is measured against a TTM revenue base that itself grew 65% last year. If FY2026 guidance is missed materially — say product revenue of $4.0bn rather than the guided $4.9–5.3bn — the demonstrated CAGR falls and the base falls together, and the margin narrows from both ends. That is the falsifiable condition and it is dated: the FY2026 outcome is known by February 2027.

Part C — The 12-month target

Built the way the horizon demands, per valuation.md: near-term estimates, named product-cycle events, and the multiple anchored on ALNY's own trading range with the percentile stated. Not a DCF. Not a peer median projected forward.

C1 — The near-term revenue base

FY2026 is management's own reiterated guidance (2026-02-12, reaffirmed 2026-04-30):

FY2027 is a house estimate — not guidance, not consensus. Product revenue +30% to ~$6,630m (TTR continuing to work the diagnosed pool, Rare flat-to-up, ONPATTRO fully cannibalised) plus $400m of collaborations and royalties = ~$7,000m.

The multiple is applied to the TTM revenue that will be reported twelve months from now — i.e. TTM through Q2-2027 ≈ (FY2026 $5,550m + FY2027 $7,000m) / 2 = $6,300m.

C2 — Named product-cycle events inside the 12-month window

Each of these appears in ALNY_Catalyst_Calendar.md with its date and source. None is invented.

1. Q2-2026 results — the first read on whether the +64–77% product guidance holds after a Q1 that delivered $1,036m (about 20% of the guidance midpoint, requiring acceleration through the year).
2. H2-2026 Phase 1 ALN-2232 obesity data — the only asset in the pipeline capable of changing the narrative multiple rather than the numbers.
3. H2-2026 Phase 1 ALN-HTT02 (Huntington's) and Phase 1/2 ALN-6400 (HHT) data.
4. Regeneron's cemdisiran NDA — submitted to FDA; additional global filings planned in 2026. An approval converts a collaboration into a royalty stream.
5. H1-2026 cAPPricorn-1 enrolment completion and Alzheimer's Phase 2 initiation for mivelsiran.

C3 — The multiple, anchored on ALNY's own history with the percentile stated

EV/TTM-sales computed daily from 2022-01-01, using time-varying share counts and net debt by year (not a constant-share approximation), against the TTM revenue reportable at each date with a 45-day filing lag:

Base multiple chosen: 10.5x — approximately the 9th percentile of ALNY's own 4.5-year distribution and below its own 2026-YTD median of 11.49x. This assumes no re-rating whatsoever toward the name's own history. It is a deliberately conservative anchor and it is stated as such.

C4 — The target

12-month target: $481, +66.7% above spot.

Read this correctly. The target is arithmetic, not optimism. TTM revenue rises about 47% on management's own reiterated guidance while the multiple is held near the bottom decile of the name's own history. The bear case — a multiple below anything ALNY has ever traded at — still produces +35%, because the denominator is growing that fast. The risk to this target is the revenue (a guidance miss), not the multiple. If FY2026 product revenue lands at $4.3bn instead of $5.1bn, the base case falls to roughly $410.

valuation.md records that 16 of 16 prior house targets sat below spot, a median 46.1% below Street, and that this expressed a house view about the market rather than a valuation. A target above spot here is the expected and normal output, and it survives a bear multiple that has no historical precedent.

C5 — Sanity band against an external professional target

Not computable. No paid feed is connected, and Alpha Vantage's quota — the free route to consensus — was exhausted by parallel agents at run time (verified). The gap to a professional's target is stated as unavailable rather than estimated. This is the correct use of the rule: an outside target is a check on the output, never a calibration target for the model, and its absence weakens the check without invalidating the output.

Part D — Reconciliation to the screen