SpaceX (SPCX) Report Interpretation

Overview

This update reviews six August developments at SpaceX. Bernstein argues that the Louisiana launch-site plan, progress toward reusable Starship operations, and continued expansion across connectivity and compute support its long-term valuation framework, while execution timing and supply constraints remain important risks.

Core views

Bernstein identifies launch cadence as the central driver of SpaceX’s value because the company’s planned orbital data centers require very large numbers of Starship launches. The proposed Starbase Louisiana on Pecan Island is intended to add major capacity: the 125,000-acre site is planned to include propellant production, power generation, deep-water shipping, vehicle processing and an airport, with at least 10 launch pads and a stated eventual capability of more than 30 Starship flights per day. Construction is expected to begin in 2027 and the first Starship launch is expected in 2029. SpaceX has committed at least $100 billion of investment over time and at least 3,000 jobs. Bernstein views the new pads as meaningful infrastructure for orbital data centers, but notes substantial work remains before the site becomes usable. The report contrasts the company’s ambitions with Bernstein’s more cautious launch assumptions. It expects five pads to be operational by end-2027—two in Texas and three in Florida—and estimates that 15 pads running at three launches per pad per day could support well above 15,000 launches annually once fully operational. If half of Louisiana’s pads were available in 2031 at that rate, they could support about 10,000 launches, consistent with the company’s projection of roughly 50GW of orbital data-center capacity launched in 2031. Bernstein nevertheless assumes roughly a one-year delay versus company launch targets and slower long-term growth, reflecting the company’s historical record of ultimately delivering but not on original timelines. Full reuse is essential: Launch 14 is expected to seek orbit and deploy operational Starlink V3 satellites, subject to FAA approval, while a second-stage catch may move to Launch 15 and actual reuse is targeted for early 2027. SpaceX’s planned Texas foundry for industrial gas-turbine blades and vanes is framed as an effort to reduce a power-buildout bottleneck. The company aims to place the power component for 20GW of terrestrial compute capacity by end-2027, but Bernstein is skeptical that new high-end blade capacity can materially affect that timetable over the next 18 months. The market is concentrated in Howmet and Precision Castparts, and the report expects SpaceX’s capacity to be for internal use rather than a broader turbine-component business. A rushed buildout, in Bernstein’s view, would prioritize timing over economics and would be unlikely to match established suppliers on cost, scale and quality. On AI, the report sees potential upside but leaves the outcome uncertain. SpaceX completed its acquisition of Cursor to add coding capability alongside Grok, while emphasizing Grok 4.6 improvements. OpenAI has said it intends to withdraw Cursor’s access to its model, whereas Anthropic said it would continue allowing Cursor to use Claude; Cursor and OpenAI remain in discussions. Bernstein’s base assumption is that orbital data centers primarily sell raw compute capacity, but better Grok capabilities and Cursor could allow additional monetization through intelligence, potentially adding revenue and margin if the strategy develops successfully. Terafab in Grimes County, Texas is another long-term compute initiative. SpaceX, Tesla and the Texas governor announced the facility on August 6; its eventual ambition is tied to a roughly 1TW compute target, although Bernstein describes this as significantly above current global supply and notes the initial phase will be far smaller. The initial build is expected to involve approximately $16.8 billion of SpaceX and Tesla capital expenditure and at least 3,000 employees. Finally, Bernstein highlights continued Starlink enterprise momentum in aviation. Royal Air Maroc joined in August and Qatar Airways expanded its rollout, bringing Bernstein’s count to 46 airline sign-ups. The report argues Starlink is the leading inflight-connectivity offering and notes that Delta and JetBlue are the only exceptions, having selected Amazon’s LEO service. Bernstein compares Starlink’s more than 11,000 satellites in orbit with 396 for LEO and expects Starlink’s airline presence to keep expanding.

Analysis framework

Bernstein reviews six recent company developments, then links each to the operating requirements behind its long-term thesis: launch cadence and reuse for orbital data centers, power and semiconductor capacity for compute expansion, AI capability for monetization, and Starlink adoption for connectivity growth. It applies deliberately slower launch assumptions than management’s targets and values the company on long-dated segment economics.

Methodology notes

• Valuation methodsEV/EBITDA valuation

  12-month forward EV/EBITDA applied to 2031 EBITDA

  Bernstein uses a long-dated EBITDA multiple because it believes nearer-term earnings would omit the orbital data-center business that is central to its valuation.

• Valuation methodsSOTP (Sum-of-the-Parts) Valuation

  Sum-of-the-parts valuation by segment

  The report values the segments using 2030 EV/EBITDA based on 2031 values, then discounts them back to mid-2027 to derive the one-year $248 target price.

• Industry AnalysisSupply-demand framework

  Launch capacity, reusability, turbine-blade supply and semiconductor availability

  The analysis tests whether launch and supply-side capacity can support the planned expansion of orbital and terrestrial compute.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• SpaceX (SPCX)

  Primary covered company; launch scale, orbital data centers, Starlink expansion and AI monetization are the report’s principal value drivers.

  Strengths

  Planned launch-pad expansion, Starlink airline adoption, large-scale compute initiatives, and potential AI monetization through Grok and Cursor.

  Weaknesses

  Launch timing assumptions are more conservative than management’s targets; the turbine-blade foundry may not affect the near-term power buildout.

  Comparison

  Bernstein states Starlink has more than 11,000 satellites versus 396 for Amazon’s LEO service and considers no competitor close in airline connectivity.

  Risks

  Full Starship reusability, orbital-data-center execution, semiconductor and supply-chain constraints, launch incidents, regulation, and compute-demand uncertainty.

Key data

• Price target$248Bernstein reiterates Outperform.
• Close price$143.69As of 31 Aug 2026; the report shows 73% implied upside.
• Starbase Louisiana footprint125,000 acresPlanned site with at least 10 launch pads and eventual capacity above 30 flights per day.
• Louisiana investment and jobsAt least $100 billion and at least 3,000 jobsSpaceX commitments over time for the new facility.
• Orbital data-center launch capacityRoughly 50GW in 2031Company projection cited by Bernstein; its launch profile is more conservative.
• Terrestrial compute power target20GW by end-2027Company target cited in the foundry discussion.
• Terafab initial capital expenditureApproximately $16.8 billionExpected SpaceX and Tesla investment for the initial phase.
• Airline Starlink sign-ups46 airlinesBernstein’s count after Royal Air Maroc joined and Qatar Airways expanded.
• Starlink satellitesMore than 11,000Compared with 396 satellites for Amazon’s LEO service, according to the report.
• 2026E revenue$46,402 millionBernstein estimate in the financial forecasts.
• 2027E adjusted EBITDA$85,583 millionBernstein estimate in the summary financial table.

Impact & implications

Bernstein argues that Louisiana’s additional launch infrastructure strengthens the path to the launch volumes required for orbital data centers, but its valuation remains dependent on Starship reuse and a slower-than-management execution ramp. It sees Starlink’s airline growth as an operating positive and AI initiatives as potential incremental monetization, while treating the foundry primarily as an internal supply-chain solution rather than a new external business.

Risks

• SpaceX may not be able to expand Starship launch capacity as required, since the ramp depends on full reusability.
• Orbital data centers may not be operationalized as assumed in the valuation.
• Semiconductor availability and other supply-chain constraints could limit execution.
• Launch incidents or regulatory restrictions could disrupt the planned ramp.
• Compute demand could be more limited than expected.

What to watch

• FAA approval for Launch 14 to reach orbit and deploy operational Starlink V3 satellites.
• Whether Launch 15 can achieve a second-stage catch and whether actual reuse begins in early 2027.
• Progress on Louisiana construction, launch-pad availability and Starship launch cadence.
• Whether the turbine-blade foundry can contribute to the planned 20GW terrestrial-compute power buildout.
• The outcome of Cursor’s model-access discussions with OpenAI and continued availability of Anthropic’s Claude.
• Further Starlink airline wins and rollout expansion.