Orbital Sovereignty: Two Kinds in One Week
On 8 September 2026, Rocket Lab released a solar cell that removes a single element from its supply chain: germanium, the mineral multi-junction spacecraft power cells have used for three decades, and the same mineral China placed under an export licence to the United States eleven months earlier. Eight days before that, on 1 September 2026, the Royal New Zealand Air Force and the Australian Defence Force's Space Command stood in Wellington and endorsed a shared Space Vision Statement, a bilateral framework built on personnel interchangeability and capability integration rather than on removing a dependency at all. Both events are New Zealand's own. Both are described, in the language each side chose, as sovereignty. They are not the same kind.
The pairing is unusual for this series. The two most recent instalments were driven entirely by developments outside New Zealand: a Shanghai municipal government's industrial-cluster designation, and an unreconciled American hardware partnership's own numbers. This week, for the first time since a Māhia launch in July, both of the week's driving stories start at home. One retires a dependency through a company's own manufacturing decision. The other obtains capability by sharing it with a larger partner. Reading them side by side is less about which strategy wins and more about which lever New Zealand is actually able to pull itself, and which one it can only join.
What happened this week
Rocket Lab's IMM Apex solar cell entered production on 8 September 2026. The company states 31.5 per cent beginning-of-life conversion efficiency, 40 per cent lower cell mass than a conventional multi-junction cell, and best-in-class power output per kilogram while retaining the radiation hardness and temperature tolerance spacecraft power systems require. The cell is offered as a mechanical and electrical drop-in replacement for the germanium-substrate cells that have powered satellites for three decades, requiring no re-tooling on the buyer's side, and Rocket Lab says it is already manufacturing in multi-hundred-kilowatt volumes. The line builds on Rocket Lab's existing IMM cell family, which has flown on NASA's Ingenuity Mars Helicopter and on satellites for more than a decade. Brad Clevenger, President of Rocket Lab USA, put the commercial case plainly: IMM Apex, he said, delivers exceptional performance while addressing rising material costs and supply chain constraints.
Eight days earlier, in Wellington, the Royal New Zealand Air Force and the Australian Defence Force's Space Command endorsed a shared Space Vision Statement following the two countries' inaugural space talks. Air Commodore Andy Scott signed for the RNZAF and Major General Greg Novak signed for the ADF's Space Command; the statement is named here once, factually, because that is what happened, not as the basis for anything that follows. Its stated objectives are personnel interchangeability, capability integration, operational readiness and a shared domain approach to space operations, framed by both parties as an extension of the wider Anzac defence relationship into the space domain specifically. Scott's own description of the RNZAF's recent history sets the scale: New Zealand stood up its first dedicated space unit, No. 62 Squadron, only last year, with a focus on understanding space activity. The precise 2025 date has not surfaced in the public record and should not be guessed at; the fact that the unit is barely a year old is the more useful data point.
Two ways to hold a dependency, not two grades of it
Space Mafia's opening chapter, on the private actors reshaping the new space economy, is the more direct anchor for the Rocket Lab half of this week's pairing than any framework built around state actors. IMM Apex is a New Zealand-founded company's own engineering response to a supply-chain exposure, made without asking anyone's permission beyond its own production line. The book's fourth fault line, inequality and consolidation, describes what happens when a critical input concentrates in too few hands; a mineral supply chain governed by one country's export-licensing regime is a reasonable extension of that fault line into hardware, though it is an extension this series has not made explicit before now and should be read as one, not as a mapping the book's own text has already confirmed.
Every prior instance of this fault line in the series has concerned compute or connectivity: who owns the GPUs, who owns the constellation, who owns the ground segment. IMM Apex is the first time a materials input, rather than a platform or a service, has supplied the example. The mechanism is the same even though the layer is different: a critical resource concentrated in few hands becomes a lever any one of those hands can pull, whether the resource is orbital compute capacity or a refined mineral.
The mineral itself is not a settled problem. China banned germanium exports to the United States outright in December 2024. Following the trade discussions between the two governments in November 2025, Beijing suspended that ban, rather than lifting it, moving gallium, germanium and antimony to a licensing regime instead of a prohibition. Exports to military end-users remain banned throughout; only civilian use may proceed, and only under licence. The suspension itself runs out on 27 November 2026 unless renewed. Rocket Lab's germanium-free cell is being marketed into that live, time-limited arrangement, not into a resolved one, and nothing in this week's findings says what happens once the licence expires.
China's actual share of global germanium production will not resolve into one number from the sources available this week. Figures from 60 to 83 per cent circulate, depending on whether a report measures refined or primary production and which year it uses as its base. Every source agrees the concentration is large; none of them agrees on a figure precise enough to headline, and presenting one as settled would overstate what is actually known. The honest description is a large majority, measured differently by different people, not a specific per cent.
Rocket Lab's own chief executive is the best available check on how far the materials story reaches. Peter Beck was asked, on the company's Q2 2026 earnings call, held 10 August 2026, whether Rocket Lab is moving toward building orbital data centres itself. He called it a real opportunity, said the company is taking it seriously and has already released solar cells built for data-centre applications, then drew the line plainly: Rocket Lab going out and building a whole lot of data centres, he said, is not something he sees happening yet. IMM Apex is a components company retiring a materials dependency, not an operator entering orbital infrastructure. That distinction matters more than the headline.
Set the two developments beside each other and a structural difference, not a ranking, is what shows up. IMM Apex removes a specific, licensed and time-limited mineral chokepoint through Rocket Lab's own engineering; the company needed no partner and no permission beyond its own production line to do it. The RNZAF-ADF Space Vision Statement obtains capability and legitimacy through interchangeability and integration with a larger ally; by its own design, that value depends on the partnership continuing to hold. Neither mechanism is shown here to be the stronger form of sovereignty. They are different in kind. One is exercised alone; the other only works for as long as someone else keeps agreeing to it.
The New Zealand position: further ahead than the standard line suggests
New Zealand's cloud-sovereignty position needs a correction before any board takes it as read. This series, and the standing guidance it has drawn on, has described New Zealand as dependent on offshore cloud capacity with the nearest hyperscaler region in Sydney. That framing is out of date. Amazon Web Services launched a dedicated Asia Pacific (New Zealand) Region, three availability zones, on 2 September 2025, with a stated NZ$7.5 billion investment; this series first noted the launch as a supporting fact at Article #16 and returns to it here because the framing built on top of it has not caught up with the fact itself. Prime Minister Christopher Luxon called it the largest publicly announced technology investment in New Zealand and said it would create up to 1,000 jobs; that GDP estimate is his own public statement, not an independently modelled figure this desk has verified, and nothing here evaluates whether the accompanying government memorandum of understanding was the right instrument. What changes is the question worth asking. It is no longer whether New Zealand's cloud infrastructure sits offshore. It now sits in-country, under a foreign hyperscaler's ownership and jurisdiction, a different and arguably more interesting question than the one the old framing kept asking.
The same two-mechanism lens this week's pairing supplies is directly useful to a board weighing its own supply-chain and platform exposure, well beyond orbital hardware. Some dependencies can be retired unilaterally, the way Rocket Lab retired a mineral input: redesign, re-engineer, or re-source, and the fix belongs to the organisation that made it. Others can only be pooled: shared standards, shared infrastructure, a consortium or an alliance, where the benefit depends on every party continuing to show up. Confusing the two is where governance goes wrong. A pooled dependency treated as though it were owned outright is a board's risk register understating what it actually controls; a unilateral fix mistaken for permanent leaves an organisation unprepared for the next licensing regime, the next export control, the next supplier that changes terms.
A cloud contract renewed on the assumption that a hyperscaler's regional presence is permanent is one instance of the pooled kind; a vendor relationship an organisation could, in principle, replace with an in-house build or a second supplier is the other. Neither assumption is safe to make from the outside. Both require the same discipline this week's pairing illustrates: ask whether the fix belongs to you, or whether it belongs to whoever you are relying on.
Two gaps are worth stating rather than filling. No source located this week says whether the New Zealand government holds any position connecting critical-minerals sovereignty to Rocket Lab's announcement specifically, or to supply-chain sovereignty as a policy area more broadly; that silence is a plausible follow-up question, not a finding this week's evidence supports. And the exact 2025 date No. 62 Squadron was established has not surfaced in the public record; last year is what Air Commodore Scott said, and it stands as stated rather than as a date this desk has independently confirmed.
What to watch next
The book's Heaven Vector and Skynet Vector framing separates what is verifiable from what merely looks reassuring, and this week supplies a clean instance of both readings sitting on the same set of facts. Read as Heaven Vector, IMM Apex is a highly legible form of sovereignty: any buyer can test whether the cell actually contains germanium, and the claim either holds or it does not. The Space Vision Statement is transparent in the same sense; it names its own objectives publicly rather than leaving them to be inferred. Read as Skynet Vector, a dependency retired inside one company's proprietary manufacturing process is legible only for as long as that company keeps disclosing it, and an alliance's practical value rests on commitments that, by their nature, are tested only in circumstances the public statement does not describe. Both readings are available from the same evidence. The book's own preference is to hold both rather than collapse them into a verdict, and this week is a genuine case for doing exactly that: neither side of the comparison has produced enough independently checkable detail yet to say which vector it is actually on. A second, independently sourced supplier confirming IMM Apex's germanium-free claim at scale would move that side of the comparison further toward Heaven Vector; a second bilateral commitment of the same kind, tested under actual operational pressure rather than announced in a joint statement, would do the same for the other.
Three dates and one dispute carry this story past this week. Whether China renews or tightens the gallium, germanium and antimony licensing regime before it lapses on 27 November 2026 will settle whether Rocket Lab's timing was fortunate engineering or fortunate positioning; nothing in this week's evidence says which. Whether a second, reconciled figure for China's share of global germanium production ever emerges, or whether the 60-to-83-per-cent range simply persists as the accepted description, is worth tracking the way this series has tracked other unreconciled numbers before. Whether the New Zealand government states any position connecting critical-minerals supply chains to sovereign capacity, in any policy area, would close the one gap this week's research could not. And whether No. 62 Squadron's actual 2025 founding date surfaces anywhere in the public record is a small fact worth having, if only because a unit barely a year old is doing more of the talking on New Zealand's space posture than its age would suggest.
The open-source dimension sits closer to the ground segment than either headline story, and it is worth naming before the defence implication, because it shows the same substitution-versus-pooling choice at a smaller, more inspectable scale. AMSAT, the Radio Amateur Satellite Corporation, has spent decades showing that an openly published ground-station design tradition can be reproduced and verified by anyone willing to build it, rather than licensed from a single vendor. NASA's Jet Propulsion Laboratory publishes its own flight software framework, F Prime, under an Apache licence; any operator, allied or commercial, can inspect precisely what it does rather than take a manufacturer's word for it. Neither project resolves this week's mineral-licensing question or the alliance's future performance. Both demonstrate a third option beside unilateral substitution and alliance pooling: infrastructure built openly enough that anyone can verify it, rather than owned outright or dependent on a partner's continued goodwill.
The RNZAF-ADF Space Vision Statement did not create New Zealand's alliance-space relationships from nothing; it extends a lattice that already exists. New Zealand has belonged to the Combined Space Operations Initiative, the multinational military space information-sharing framework, since before this statement was signed, alongside Australia, the United States, the United Kingdom and several other allied states. Allied early-warning architecture such as the Space-Based Infrared System shows the same pooling logic at a different layer: sensor coverage and warning time obtained by sharing a capability no single small state would build alone. None of this describes New Zealand acquiring a unilateral capability the way Rocket Lab's own manufacturing decision does. It describes New Zealand deepening its place inside a framework whose value, like the Space Vision Statement's own, depends entirely on the other members continuing to hold up their side of it.
Where has your organisation mistaken a pooled dependency, a shared platform, a consortium standard, an alliance of vendors, for one it actually owns outright, and where has it done the opposite: sat on a fix it could make alone while waiting for a partner who was never going to move first? I would like to hear about it.
This is one of seven weekly series in The Hamberger Report. Subscribe on LinkedIn and the next one arrives in your feed.
The views expressed in this article are entirely my own, informed by more than 30 years of professional experience in architecture, security, and technology leadership in New Zealand. I write as director of Te Pono Limited; the views are personal and do not represent the position of any client, any government agency, or the New Zealand government. My commentary on legislation and policy is analytical, drawing on publicly available sources and my professional expertise in architecture, security, and AI governance, and it is politically neutral.
Andreas Hamberger is a New Zealand leader in Architecture & Security and Associate Member of the Institute of Directors. Space Mafia examines the sovereignty implications of orbital compute infrastructure.
This article was produced with AI assistance under my direction. Research, drafting and images pass through a pipeline I built and govern: automated gates for source verification, forbidden language and political neutrality, and my own review before anything is published. The tools include Claude, Gemini and Openart. The frameworks, arguments and editorial judgements are mine and are the same discipline I apply to the AI systems I audit for clients. AI accelerated the work; the thinking, and the responsibility for it, are mine.
[1] GlobeNewswire, via Rocket Lab. "Rocket Lab Introduces High-Efficiency Solar Cell to Reduce Reliance on Supply-Constrained Critical Minerals." 8 September 2026. https://www.globenewswire.com/news-release/2026/09/08/3358129/0/en/rocket-lab-introduces-high-efficiency-solar-cell-to-reduce-reliance-on-supply-constrained-critical-minerals.html
[2] New Zealand Defence Force. "New Zealand, Australian militaries move to formalise defence collaboration on space." 1 September 2026. https://www.nzdf.mil.nz/media-centre/news/new-zealand-australian-militaries-move-to-formalise-defence-collaboration-on-space/
[3] Australian Aviation. "RNZAF and Australian Defence Force endorse shared Space Vision Statement." September 2026. https://australianaviation.com.au/2026/09/rnzaf-and-australian-defence-force-endorse-shared-space-vision/
[4] GlobalSecurity.org. Corroborating coverage, RNZAF-ADF Space Vision Statement. September 2026.
[5] The Oregon Group. "China Lifts Export Ban on Gallium, Germanium, and Antimony." 10 November 2025. https://theoregongroup.com/commodities/gallium-germanium/china-lifts-export-ban-gallium-germanium-antimony/
[6] Rocket Lab, via Investor Relations. "Rocket Lab Announces Second Quarter 2026 Financial Results." 10 August 2026. https://investors.rocketlabcorp.com/news-releases/news-release-details/rocket-lab-announces-second-quarter-2026-financial-results-posts
[7] The Motley Fool. "Rocket Lab (RKLB) Q2 2026 Earnings Call Transcript." 17 August 2026. https://www.fool.com/earnings/call-transcripts/2026/08/17/rocket-lab-rklb-q2-2026-earnings-call-transcript/
[8] Reseller News. "AWS launches sovereign-by-design New Zealand cloud region." September 2025. https://www.reseller.co.nz/article/4049313/aws-launches-sovereign-by-design-new-zealand-cloud-region.html