Ads and analytics load only if you accept.Ads and analytics cookies load only if you accept. Reject and you get no ads at all. Essential cookies keep the site working.PolicyCookie Policy ·
Newsletter
Week in Space Data, every Friday.
Quarterly release · Q3 2026
SpaceNexus Supply-Chain Concentration Report
Where the tracked space supply chain concentrates: by country, by tier, and on the suppliers most programmes depend on.
Share of mapped relationships, not of the industry
Single-source dependencies
10
4 elevated-risk components also name one supplier
Concentration by supplier country
Mapped supply relationships attributed to the supplier’s country of domicile.
Concentration by supplier country. Mapped supply relationships attributed to the supplier’s country of domicile. Showing the first 10 of 12 rows.
#
Country
Suppliers
Relationships
Share of map
Critical
High geopolitical risk
1
United States
13
15
+50.0%
13
0
2
France
3
3
+10.0%
3
0
3
Europe
2
2
+6.7%
1
Concentration by supplier tier
Where in the chain the mapped supply sits.
Concentration by supplier tier. Where in the chain the mapped supply sits. 4 rows.
Tier
Companies mapped
Relationships supplied
Share of map
Critical
Tier 1 supplier
21
16
+53.3%
15
Tier 3 supplier
16
8
+26.7%
6
Tier 2 supplier
18
6
+20.0%
5
Prime contractor
8
0
0.0%
0
Chokepoints: suppliers by dependent customers
Suppliers ranked by how many distinct mapped customers depend on them. The fan-in figure, not a revenue figure.
Chokepoints: suppliers by dependent customers. Suppliers ranked by how many distinct mapped customers depend on them. The fan-in figure, not a revenue figure. Showing the first 10 of 27 rows.
#
Supplier
Country
Tier
Dependent customers
Critical relationships
1
Aerojet Rocketdyne
United States
Tier 1 supplier
2
2
2
BAE Systems Electronic Systems
United Kingdom
Tier 1 supplier
2
2
3
Honeywell Aerospace
United States
Tier 1 supplier
2
2
4
ArianeGroup
Single-source critical dependencies
Mapped customers for which exactly one mapped supplier provides a relationship marked critical.
Single-source critical dependencies. Mapped customers for which exactly one mapped supplier provides a relationship marked critical. 10 rows.
Customer
Sole critical supplier
Supplier country
Products
Aerojet Rocketdyne
ATI Specialty Materials
United States
nickel_superalloys; titanium
Analog Devices
TSMC
Taiwan
advanced_semiconductors
ArianeGroup
Safran
France
vulcain_engine_components
ATI Specialty Materials
Glencore (Cobalt)
Switzerland
cobalt_for_superalloys
Ball Aerospace
BAE Systems Electronic Systems
United Kingdom
rad_hard_processors; space_electronics
Honeywell Aerospace
China Northern Rare Earth
Elevated bill-of-materials risk by category
Components in the risk register at critical or high risk, grouped by subsystem category.
Elevated bill-of-materials risk by category. Components in the risk register at critical or high risk, grouped by subsystem category. 7 rows.
Category
Critical
High
Total
avionics
2
0
2
communications
1
2
3
life_support
1
2
3
structure
1
2
3
payload
1
1
2
thermal
1
0
1
Elevated-risk components
Every component the register rates critical or high, sole-supplier items first.
Elevated-risk components. Every component the register rates critical or high, sole-supplier items first. Showing the first 10 of 17 rows.
Recorded shortages whose start date falls on or before the end of the quarter and for which no resolution is recorded.
Shortages active at 2026-09-30. Recorded shortages whose start date falls on or before the end of the quarter and for which no resolution is recorded. Showing the first 10 of 40 rows.
Material
Category
Severity
Started
Estimated resolution
Companies impacted
Advanced Composites (CFRP)
structures
medium
–
2025
3
Aerospace-Grade PAN Carbon Fiber Precursor
structures
high
–
2027
5
Aerospace-Grade Titanium
specialty_metals
high
–
2026
3
Beryllium
specialty_metals
high
–
Methodology
•Computed over the analyst-maintained supply-chain map (companies, relationships and shortages) and the bill-of-materials risk register. Both are reviewed rather than scraped, so they change in steps, not continuously.
•Concentration shares are counts of mapped relationships, not dollars or volumes. A country holding 30% of mapped relationships holds 30% of the relationships WE HAVE MAPPED, which is not the same as 30% of the industry.
•A relationship is "critical" when the register marks it so. Criticality is an analyst judgement recorded in the data, and it is the only judgement in this release.
•A single-source dependency is a customer for which exactly one mapped supplier supplies a relationship marked critical. Suppliers outside the map cannot be counted, so this is an upper bound on single-sourcing we can see and a lower bound on the real figure.
•Shortages are counted as active at the end of the quarter when their recorded start date falls on or before that date and no resolution is recorded.
What this edition can and cannot see
Coverage limits travel with the numbers, here and in every export of them.
•Computed over a curated map of 63 companies and 30 relationships, plus a register of 48 bill-of-materials items. Both are analyst-maintained and reviewed rather than scraped.
•Every share on this page is a share OF THAT MAP. A country holding a third of mapped relationships holds a third of what we have mapped, which is not a market share and must not be cited as one.
•Relationships are attributed to the supplier’s country of domicile, not to where manufacturing physically happens. A supplier producing in a third country is counted under its domicile.
•Criticality and risk level are analyst ratings recorded in the data. They are the only judgements in this release; every other figure is a count or a ratio over those records.
•The map has no time series. It changes when an analyst revises it, so consecutive editions can be identical, and when they are, this page says so instead of presenting unchanged figures as movement.
•Nothing on this page is model-generated.
Sources and licences
The same credits are written into every CSV and JSON export of this data, so a downloaded file still names its sources.
UK Companies House: Company registration data from the UK Companies House register. Contains public sector information licensed under the Open Government Licence v3.0. Open Government Licence v3.0
Cite this edition
Cite or embed this data
Citation
SpaceNexus Supply-Chain Concentration Report, Q3 2026. SpaceNexus, data as of 2026-09-30. https://spacenexus.us/releases/supply-chain-concentration/2026-Q3. https://spacenexus.us/releases/supply-chain-concentration/2026-Q3 (retrieved 2026-10-01).
Free to cite and embed with attribution. Data updates continuously; the citation date pins your retrieval.
SN
By SpaceNexus Data
Computed from SpaceNexus data - no model is involved in producing a release
Every figure here is arithmetic over our own rows (counts, sums, deltas and rankings), and the page and its export are built by the same code, so they cannot disagree. How this edition is computed
0
4
United Kingdom
1
2
+6.7%
2
0
5
Canada
1
1
+3.3%
1
0
6
China
1
1
+3.3%
1
1
7
Germany
1
1
+3.3%
1
0
8
Japan
1
1
+3.3%
1
0
9
Norway
1
1
+3.3%
0
0
10
Russia
1
1
+3.3%
1
1
Shares are of the 30 relationships in our map. They are not market shares and must not be cited as such.
Showing the first 10 of 12 rows. 2 more in this table, plus every column the display leaves out.
The complete row set and its CSV and JSON exports are part of a SpaceNexus Research seat. The figures above are computed from exactly the same rows, so the summary and the export can never disagree. What a Research seat covers.
France
Tier 1 supplier
1
1
5
ATI Specialty Materials
United States
Tier 3 supplier
1
1
6
Azur Space Solar Power
Germany
Tier 2 supplier
1
1
7
China Northern Rare Earth
China
Tier 3 supplier
1
1
8
Cobham Advanced Electronic Solutions
United States
Tier 2 supplier
1
1
9
Collins Aerospace (RTX)
United States
Tier 1 supplier
1
1
10
Glencore (Cobalt)
Switzerland
Tier 3 supplier
1
1
Showing the first 10 of 27 rows. 17 more in this table, plus every column the display leaves out.
The complete row set and its CSV and JSON exports are part of a SpaceNexus Research seat. The figures above are computed from exactly the same rows, so the summary and the export can never disagree. What a Research seat covers.
Suppliers outside the map cannot be counted, so this is an upper bound on the single-sourcing we can see and a lower bound on the real figure.
propulsion
0
3
3
3
24-36 months (custom ASIC), 12-18 months (FPGA)
Lunar Regolith-Compatible Seals
structure
critical
3
4
36-60 months (R&D + qualification)
Rad-Hardened SRAM / DRAM
avionics
critical
3
2
12-18 months
Zero-Boil-Off Cryocoolers (Large-Scale)
thermal
critical
3
3
36-48 months (development + qualification)
Closed-Loop Water Recovery System
life_support
high
1
3
24-36 months (development + qualification)
CO2 Removal Canisters (CDRA-type)
life_support
high
1
3
12-18 months
Green Propellant (AF-M315E / LMP-103S)
propulsion
high
2
3
9-15 months
Risk level is an analyst rating recorded in the register. It is the only judgement on this page; every other figure is a count.
Showing the first 10 of 17 rows. 7 more in this table, plus every column the display leaves out.
The complete row set and its CSV and JSON exports are part of a SpaceNexus Research seat. The figures above are computed from exactly the same rows, so the summary and the export can never disagree. What a Research seat covers.
Ongoing
3
Closed-Loop Water Recovery Systems
life_support
high
–
2029
4
Cobalt
specialty_metals
medium
–
2026
3
Deep Space Transponders (X-Band/Ka-Band)
communications
medium
–
2027
6
ECLSS CO2 Removal Canisters (LiOH/Metal Oxide)
life_support
high
–
2028
5
EVA Suit Components (Next-Generation)
robotics
critical
–
2029
4
Gallium
semiconductors
high
–
2027
3
40 recorded shortages carry no start date and are treated as standing. They are counted in every edition until the register dates or closes them.
Showing the first 10 of 40 rows. 30 more in this table, plus every column the display leaves out.
The complete row set and its CSV and JSON exports are part of a SpaceNexus Research seat. The figures above are computed from exactly the same rows, so the summary and the export can never disagree. What a Research seat covers.
•
Each edition stores a hash of its inputs. When the hash matches the previous edition, the page states plainly that nothing in the map changed: an unchanged quarter is reported as unchanged rather than dressed up.