All research

Equity research

SpaceX’s Casting Plan Took $7.9bn Off Howmet for Under 1% of Its Revenue

Howmet casts more than half of the world’s industrial gas turbine blades, and its gas turbine revenue grew 41% in the first nine months of 2026. The day SpaceX’s casting plan broke took $7.9bn off its market value, for under 1% of revenue at risk, and new capacity takes even Howmet two years. We initiate Overweight on Howmet (PT $238) and Siemens Energy (PT €150), Neutral on GE Vernova (PT $925) and Underweight on Mitsubishi Heavy (PT ¥3,100).

By K2 Research TeamOct 29, 2026PDF · 13 pages
SpaceX’s Casting Plan Took $7.9bn Off Howmet for Under 1% of Its Revenue

Howmet casts more than half of the world’s industrial gas turbine blades, and its gas turbine revenue grew 41% in the first nine months of 2026. The day SpaceX’s casting plan broke took $7.9bn off its market value, for under 1% of revenue at risk, and new capacity takes even Howmet two years. We initiate Overweight on Howmet (PT $238) and Siemens Energy (PT €150), Neutral on GE Vernova (PT $925) and Underweight on Mitsubishi Heavy (PT ¥3,100).

What’s inside

  1. SpaceX can cast a blade, not a backlog: Overweight Howmet and Siemens Energy
  2. GE Vernova beat, raised and fell
  3. What the market heard: 11 GW
  4. The blade claim holds, with a caveat
  5. Orders doubled; capacity did not
  6. Lead times have stretched to five years
  7. The risk is price, not volume
  8. Demand: the grid queue sends data centres to gas
  9. Four casters, one leader
  10. The blade is cheap relative to the delay
  11. Credible intent, no capacity yet
  12. Siemens Energy is the discount; Howmet has de-rated
  13. Results days pay; the weeks after no longer do
  14. Why the pair, and why now
  15. Price target calculation and key risks
  16. What would make us wrong?
  17. Signposts

More research

View all

Every New Dollar of Backblaze Revenue Needs $2 of Drives and Servers
Equity researchPDF · 8 pp

Every New Dollar of Backblaze Revenue Needs $2 of Drives and Servers

Backblaze’s margins have never been higher and B2 growth has re-accelerated to 47%, yet the stock is 59% below its August high after October’s AI-infrastructure sell-off. The catch is the cash: each new dollar of ARR needs about $2 of drives and servers, bought from makers that have raised prices per TB by 28%, and we expect free cash flow of -$38M in 2027. We initiate at Neutral with a Dec-27 price target of $10 (+19%).

K2 Research TeamNov 7, 2026
Reddit Didn’t Sink Alphabet, Its $51bn Capex Quarter Did
Equity researchPDF · 10 pp

Reddit Didn’t Sink Alphabet, Its $51bn Capex Quarter Did

Alphabet lost about $500bn of market value on 29 October on its own 3Q26 print; Reddit reported after that close and fell about 30% the next day. What the market sold was cash flow: capex of $51.4bn exceeded operating cash flow of $42.3bn, and 2027 capex will rise significantly. Search still grew 12% on 8% more paid clicks. We initiate at Neutral with a Dec-27 price target of $310.

K2 Research TeamNov 5, 2026
Neoclouds Fell Up to 17% This Week, and Beta Explains Nearly All of It
Equity researchPDF · 9 pp

Neoclouds Fell Up to 17% This Week, and Beta Explains Nearly All of It

All four neoclouds fell more than the Nasdaq-100’s 4.8% after Microsoft’s and Alphabet’s capex misses, from SpaceX (-7.2%) to IREN (-17.0%), and beta explains each move to within 3pp. We reiterate Overweight on Nebius (PT $207, now +34% upside) and IREN (PT $37), and Underweight on SpaceX (PT $90) and CoreWeave (PT $44).

K2 Research TeamNov 1, 2026