SpaceX (Nasdaq: SPCX) — Living Context

Coverage desk context file, per content/coverage/PROCESS.md §1. Author: Opus (writing stage). As of 2026-07-15. This is the standing reference for the name — company facts, capital structure, the sources the desk relies on, open questions, and the changelog. It is revised on every material event; see the Changelog at the foot for revision history.

Tier. Tier 1 (SEC-reporting) since the 2026-06-12 Nasdaq listing. Coverage was scoped under the roster's stale Tier 2 (private) entry; the disclosure sweep found the company had already listed, and the initiation was rebuilt Tier 1 — anchored to the S-1/424B4 and 8-Ks, with the call expressed against the traded price. See filings.md §0 and PROCESS.md §7.

---

1. Company card

Ownership & control. Founder-controlled. SpaceX historically maintained a multi-class structure with founder super-voting shares (consistent with Musk's stated intent to retain control given the Starship/Mars mission profile). The post-IPO capital structure was set by the IPO-closing omnibus 8-K of 2026-06-15 (Items 3.02/3.03/5.03), which adopted new charter/bylaws and converted the pre-IPO classes into the public structure. The exact post-IPO vote structure and share-class mechanics have not yet been pulled verbatim from the 424B4 ("Description of Capital Stock" / "Principal Stockholders") — a super-voting arrangement preserving Musk's board control is the desk's working assumption but is unconfirmed (see Open Questions #2; filings.md §2e).

Share count (working figure, unconfirmed). ~13.0B shares outstanding, back-solved from $1.75T ÷ $135 IPO price — not a disclosed figure. The entire per-share price-target arithmetic rests on this number; a materially different filed count from the 424B4 cover page rescales every price-target figure proportionally and would be restated before the next coverage update (PROCESS.md §4). See Open Questions #1.

Structural note — the xAI merger. The three-segment structure exists only because of the all-stock merger with xAI Corp., closed 2026-02-02/03 at an implied combined value of $1.25T (SpaceX $1T + xAI $250B — CNBC, 2026-02-03). Before it, SpaceX was a launch-and- broadband company; the AI segment (and the AI-capex story that dominates the valuation) enters the P&L only from that transaction. It was structured as a related-party, all-stock deal.

---

2. Capital structure & valuation history at a glance

Current market (as of 2026-07-14 close). Price $136.12/share; implied market cap ≈ $1.77T (≈13.0B sh × $136.12); estimated post-IPO net cash ≈ $50B; implied EV ≈ $1.72T (model.md §6). Trailing multiple ≈ 95x FY2025 revenue of $18.67B — an outlier against every public comp (Rocket Lab ~$50.6B cap, AST SpaceMobile ~$26.2B, Planet Labs ~$9.6B; ~$86B combined — research.md §6.1).

IPO. Priced $135/share on 2026-06-11, ~$75B raised (primary + secondary combined), ~$1.75T pricing basis; listed Nasdaq: SPCX 2026-06-12; first-day close $160.95 (+19%, ~$2.0–2.1T implied cap). Stock peaked near ~$2.7T implied cap within days of listing, then round-tripped to roughly flat vs. the IPO price by mid-July (CNBC/TechCrunch/Forbes/Nasdaq, 2026-06-12; Nasdaq quote checked 2026-07-15). The largest IPO on record.

Private valuation ladder (superseded as the anchor by the traded price; retained for context). No primary capital was raised between the Jan-2023 round and the June-2026 IPO — every intervening mark was a non-dilutive secondary/tender event.

The mark ran 13x off the Jan-2023 $137B primary in ~3.5 years, with the last 18 months (Dec-24 → Jun-26) a ~5x move coincident with the AI/xAI narrative entering the story — i.e., the AI re-rating, not core-business growth, drove the last leg. That is the pattern the desk's Underweight is a discipline against.

Debt. Historically minimal; the model reads a FY24–25 bridge-debt build (to ~$19.5B) ahead of the listing, ~$15B assumed repaid from IPO proceeds, with a return to debt issuance by FY28E–FY30E as capex outruns operating cash flow (model.md §3, notes 1–3; unconfirmed against 424B4 "Use of Proceeds," Open Questions #3).

Desk call. Rating Underweight; 12-month price target $75–$115/share (base case $94), against the $136.12 last close — ~31% downside to base, 15%–45% downside across the working range. The full model span ($43 bear / $177 bull) is treated as tail scenarios, not the working range (coverage.md Valuation; model.md §6b).

---

3. Data-source registry

Every recurring source the desk relies on for this name, graded A (filing / regulatory / exchange-quote), B (major-outlet reporting), C (analyst estimate / secondary aggregation). Grade governs how much weight a figure carries and whether it must be closed against primary text before the model locks.

Standing sourcing caveat. Because no 10-K/10-Q exists yet, the S-1/424B4 is the only audited-adjacent, line-item disclosure — and many of its figures currently trace to grade-C secondary compilations, not verbatim primary text. Closing that gap (a direct EDGAR pull of S-1/424B4 primary text) is the top-priority follow-up before the actuals column is treated as locked.

---

4. Open questions (unresolved at initiation)

1. Exact post-IPO share count. ~13.0B is back-solved from $1.75T ÷ $135, not disclosed. Every per-share figure rests on it. Resolve via: 424B4 cover page; first 10-Q. Mechanical restatement trigger — a materially different count rescales the whole price target.
2. Vote structure & share classes. Super-voting founder control is the working assumption, unconfirmed. Resolve via: 424B4 "Description of Capital Stock" / "Principal Stockholders"; the 2026-06-15 8-K exhibits.
3. Use of proceeds. ~$75B earmarked across Starship, Starlink Gen2/Gen3, and AI compute per secondary sourcing; specific dollar allocations and any bridge-debt paydown unconfirmed. Resolve via: 424B4 "Use of Proceeds."
4. Capex-by-segment split. Only the FY25 AI figure ($12.7B of $20.7B) is a hard data point; the rest is desk estimate. Resolve via: a 10-Q/10-K segment footnote.
5. Anthropic Cloud Services Agreement (~$1.25B/month through May 2029). Single-sourced in the sweep, unverified, and excluded from every model input. If confirmed via primary text or Anthropic's own disclosure it becomes a first-order input to the AI-segment forecast. Resolve via: primary S-1 text; a subsequent 10-Q; Anthropic disclosure.
6. Over-allotment / greenshoe exercise & final net proceeds. Plausibly in the 2026-06-23 8-K but not confirmed verbatim. Resolve via: first 10-Q.
7. Segment-reporting stability. Whether the three-segment structure persists or subdivides (e.g., X/Grok split from compute) is unknown until a full post-IPO quarterly cycle.
8. Verbatim risk-factor text. filings.md §2f is a directional synthesis of secondary coverage, not a primary extraction. Resolve via: direct S-1 "Risk Factors" pull.
9. ~20% federal-revenue figure. Grade-B/C; individual award lines do not fully reconcile to the company-cited ~$22B backlog without a USASpending.gov pull.

---

5. Monitoring list (what to watch, and where)

Ranked by how much each moves the price target (thesis.md §3.3; coverage.md "What would change our view").

---

Changelog

2026-07-15 — Coverage initiated

Full initiation published as a Tier 1 name (re-tiered from the roster's stale Tier 2/private entry on the 2026-06-12 Nasdaq listing; tier-migration protocol exercised on day one per PROCESS.md §2/§4). Document set produced: filings.md (disclosure digest — S-1/424B4, six 8-Ks, private valuation ladder, regulatory & contract record), research.md (TAM, competitive map, tailwinds/headwinds, KPI dashboard, comps), model.md (three-statement model FY23A–FY30E, bear/base/bull scenarios, valuation bridge — balance sheet ties, cash flow ties), thesis.md (straw man / steel man / baseline with five swing factors), coverage.md (initiating-coverage report), and this context.md.

Call: Underweight · 12-month price target $75–$115/share (base case $94), against the $136.12 last close (2026-07-14, Nasdaq); implied market cap of the target range ≈ $0.98T– $1.50T vs. ~$1.77T today. The thesis is a valuation call, not a fundamentals call: launch and Starlink are dominant and profitable and grow under every scenario, but the price embeds AI-segment success the desk is not willing to underwrite at even 50/50 odds on one disclosed year of segment data (FY2025 AI: ~$3.2B revenue, –$6.355B operating loss). Actuals anchored to the S-1 (2026-05-20, amended 06-01/03) and 424B4; no 10-K/10-Q filed yet (first 10-Q, quarter ended 2026-06-30, expected ~August 2026 — the next scheduled coverage trigger). Open items at initiation: back-solved ~13.0B share count, post-IPO vote structure, use of proceeds, capex-by-segment split, and the unverified/excluded Anthropic compute agreement (§4).