The short version

• Most space law jobs are not public international law jobs. They are regulatory, contracts and export control jobs with space as the subject matter.
• The four routes that actually hire: satellite and telecoms regulatory, aviation regulatory, government contracts, and export control.
• A specialist LLM helps with credibility and network. It does not substitute for regulatory experience.
• The sector is small enough that the people who do this know each other, which makes writing and showing up unusually effective.

The honest starting point for anyone thinking about space law as a career is that the field people imagine and the field that exists are different.

The imagined version involves treaty interpretation, arguments about lunar sovereignty and appearances before international bodies. That work exists. There are perhaps a few dozen people in the world who do it as the majority of their practice, most of them academics or government legal advisers.

The version that hires is regulatory and commercial. Getting a client through an FAA launch licence, an FCC space station application or an ITU filing. Negotiating a launch services agreement. Working out whether a component is on the US Munitions List. Advising on whether a constellation will fall inside the EU Space Act. That work has grown a great deal in the last decade and is still growing.

If the second version sounds less interesting than the first, that is worth knowing before you invest in an LLM.

The four doors in

Satellite and telecoms regulatory

The most direct route. Firms with established satellite regulatory practices, mostly in Washington DC, London and Brussels, handle FCC space and earth station licensing, market access, ITU coordination through national administrations, and now the transition from FCC Part 25 to the new Part 100 adopted in July 2026.

What you need: telecoms regulatory experience, comfort with technical filings, patience for procedural detail. A background in communications law transfers almost completely.

Aviation and transport regulatory

Launch licensing is administratively closer to aviation than to anything else. The FAA’s Office of Commercial Space Transportation sits inside the FAA, uses FAA rulemaking machinery, and Part 450 reads like an aviation safety rule because it is one. The UK route is the same, with the Civil Aviation Authority as the space regulator under the Space Industry Act 2018.

What you need: aviation regulatory experience, or a willingness to learn safety case analysis and risk thresholds. We set out how Part 450 works here.

Government contracts

A large share of the money in the sector is government money. NASA, ESA, national defence procurement, and the contracts underneath them. Firms with strong government contracts practices ended up with strong space practices almost by accident.

What you need: procurement law, cost accounting standards, bid protest experience.

Export control

Spacecraft and their components sit on the US Munitions List under ITAR Category XV or on the Commerce Control List in the 9×515 series. Every hardware transaction, every foreign employee, every technical exchange runs through this analysis. It is unglamorous and consistently in demand.

What you need: ITAR and EAR experience. It is one of the few areas where junior lawyers get real responsibility quickly, because the volume is high.

Where the jobs sit

On the specialist LLM question

The established programmes are McGill’s Institute of Air and Space Law, Leiden’s International Institute of Air and Space Law, the University of Mississippi’s Center for Air and Space Law, and a growing number of newer offerings.

What they give you is real: a working knowledge of the treaty framework, the vocabulary to hold a conversation with people already in the field, a supervised piece of writing you can point to, and a network that is genuinely useful because the field is small.

What they do not give you is a licensing practice. Nobody hires a lawyer to file an FCC application because they wrote a thesis on Article II. Firms hire regulatory experience and then teach the space part.

The most reliable pattern we see is people who build a transferable regulatory skill first and add the space specialism, rather than the other way round. If you are choosing between an LLM and two years in a telecoms regulatory team, take the two years.

If you are already in practice and want the LLM for credibility and network, that calculation is different and often worth it.

What actually differentiates candidates

Three things, in order.

Technical literacy. Not an engineering degree. The ability to read a mission profile, understand what an orbit regime implies, follow a debris analysis without needing every term explained, and ask an engineer a sensible question. This is the single most commented-on difference between lawyers who thrive in the sector and lawyers who do not.

Regulatory stamina. The work involves long documents, iterative agency engagement and long timelines. A Part 450 application for a novel vehicle can take well over a year from first engagement. People who need fast closure are unhappy here.

Visible output. The field is small enough that writing gets read. A well argued case note on a decision that has just landed, or a clear explainer on a rule change, will be read by a meaningful percentage of the people who work in this area. That is not true in most practice areas and it is a genuine advantage for anyone starting out.

A practical first year

If you are starting from zero and want to be credible in twelve months, this is what we would suggest.

Read the Outer Space Treaty, then the Liability Convention, then the Registration Convention. Total reading time under two hours. Our article by article walk through covers what matters.

Read 14 CFR Part 450 and 47 CFR Part 25. Not to memorise them, but to understand the shape of a licensing regime and the distance between the treaty layer and the working layer.

Follow the live files. As at August 2026 those are the FCC’s Part 100 transition, the FAA’s environmental waiver proposal, the EU Space Act trilogue, and the COPUOS space resources principles due in 2026 and 2027. Being current on four live files makes you more useful in a conversation than knowing the treaties well.

Write something. A case note on the DISH consent decree or the Dark-Sky decision, an explainer on the five year deorbit rule, an analysis of what the EU Space Act would mean for a non-EU operator. Then send it to people. This works far better than it should.

Go where the people are. The International Astronautical Congress, the International Institute of Space Law’s events, the Manfred Lachs Moot if you are still a student, and the national space agency and regulator briefings which are usually open and usually under-attended by lawyers.

A note on where you are

The sector is concentrated but less than it was. Washington DC remains the centre of gravity for launch and satellite regulatory work. London has become the largest single source of space insurance capacity and the UK CAA is now an active launch regulator. Luxembourg built a space resources practice around its 2017 law. Bengaluru and Delhi are growing quickly since the Indian Space Policy 2023 made IN-SPACe the single window authorising body and FDI rules were liberalised in 2024. Tokyo, Sydney, Wellington and Abu Dhabi all have working regimes and therefore working practices.

If you are not in one of those places, the writing route matters more, and it works.

The short version

• The Outer Space Treaty is about 2,500 words long, entered into force on 10 October 1967, and had 118 states parties as at 1 January 2026.
• Six articles do almost all the work: I, II, VI, VII, VIII and IX.
• Article VI is the commercial one. It makes states answerable for their companies, which is why licensing regimes exist.
• The treaty settles almost nothing about resources, debris, traffic or brightness. That is not a drafting failure. Those problems did not exist in 1966.

Most explanations of the Outer Space Treaty start with the phrase “province of all mankind” and end with a vague gesture at peaceful use. That is the ceremonial reading. The working reading is narrower and more useful, because six of its seventeen articles carry essentially all the legal weight.

The treaty was adopted by UN General Assembly resolution 2222 (XXI) on 19 December 1966, opened for signature in London, Moscow and Washington on 27 January 1967, and entered into force on 10 October 1967. It was negotiated in about eighteen months by two superpowers who both wanted to stop the other one claiming the Moon.

Here is what each of the load-bearing articles actually does.

Article I: freedom of use, for everyone

Exploration and use of outer space shall be carried out for the benefit and in the interests of all countries, and outer space shall be free for exploration and use by all states without discrimination.

Two practical consequences. First, there is no permission requirement in international law to go to space. You do not apply to anyone. Second, “use” is a protected activity and it is broad. The argument that extracting resources is a form of “use” protected by Article I starts here.

The “province of all mankind” phrase sits in this article. It is aspirational language, not a property rule. It has never been read as creating a common heritage regime, which is precisely what the Moon Agreement tried to add twelve years later and why so few states joined it.

Article II: the non-appropriation rule

Outer space, including the Moon and other celestial bodies, is not subject to national appropriation by claim of sovereignty, by means of use or occupation, or by any other means.

Outer Space Treaty, Article II

One sentence, and the most litigated sentence in the field despite never having been litigated.

Note what it covers. It prohibits appropriation of the place. It says nothing explicit about material removed from that place. The drafters were worried about territorial claims of the kind that carved up Antarctica and Africa. They were not thinking about regolith processing.

That silence is the crack that four national space resources laws and Section 10 of the Artemis Accords have grown into. We set out both sides of that argument in our piece on space resources.

The other thing to notice is the phrase “by any other means”. Critics of the resources position lean on it hard. If a company operates an exclusive mining site for twenty years under a national licence, they argue, that is occupation dressed up in commercial clothing, and the Accords’ safety zones concept makes the exclusivity explicit.

Article VI: the article that pays your fees

States parties bear international responsibility for national activities in outer space, whether carried on by governmental agencies or by non-governmental entities, and the activities of non-governmental entities require authorisation and continuing supervision by the appropriate state party.

This is the hinge between international law and commercial practice.

A private company is not a party to the treaty and cannot be. So the treaty makes the state responsible for private conduct, and then requires the state to authorise and supervise it. Every national licensing regime in existence is a state discharging that obligation.

Note “continuing supervision”. The obligation does not end at grant. It is why licences carry ongoing conditions, reporting duties and end-of-life requirements, and why the FCC could treat DISH’s disposal plan as an enforceable licence condition years after the satellite launched.

Article VII: liability, in outline

Each state party that launches or procures the launching of an object into outer space, and each state party from whose territory or facility an object is launched, is internationally liable for damage caused by that object.

Four categories of launching state, and they can all be liable for the same object. A satellite built in Germany, procured by a Luxembourg company, launched from French Guiana on a US vehicle can implicate several states at once.

Article VII is the outline. The 1972 Liability Convention fills in the detail, splitting damage into absolute liability on the surface of the Earth and fault liability everywhere else.

Article VIII: jurisdiction follows registration

The state of registry retains jurisdiction and control over the object and any personnel on it, and ownership of objects is not affected by their presence in outer space or by their return to Earth.

This does two useful things. It supplies the connecting factor for criminal and civil jurisdiction on board, which is how the Intergovernmental Agreement on the Space Station allocates jurisdiction module by module. And it confirms that a satellite remains the property of its owner while in orbit, which is the foundation for satellite financing, security interests and salvage arguments.

The 1975 Registration Convention builds the machinery: launching states file object data with the UN Secretary-General. It has 77 parties, and compliance is patchy for smallsats.

Article IX: due regard, contamination and consultation

States shall conduct activities with due regard to the corresponding interests of other states, shall avoid harmful contamination of space and adverse changes in the Earth’s environment, and shall consult before proceeding with an activity that would cause potentially harmful interference.

Article IX is the closest the treaty comes to an environmental provision, and it is doing a lot of unglamorous work in modern debates. Debris mitigation guidelines, planetary protection policy and the dark and quiet skies argument all cite it.

Its weakness is that “due regard” and “potentially harmful interference” are undefined, and the consultation mechanism has never been formally triggered by one state against another over a constellation.

What the treaty does not do

It is worth being blunt about the gaps, because a lot of commentary pretends they are not there.

None of this makes the treaty a failure. It has held for nearly sixty years across a Cold War, the collapse of one of its two authors and the arrival of an industry its drafters could not have imagined. But it is a framework, and frameworks need floors built under them. Most of the interesting legal work in space today is that floor building, happening in national rulemaking and in COPUOS working groups rather than in treaty negotiation.

The short version

• Space law is a small stack of UN treaties from 1967 to 1979, sitting on top of a much larger and faster moving pile of national licensing rules.
• The Outer Space Treaty makes states responsible for what their companies do in orbit. That single sentence is why every launch and every satellite needs a government licence.
• Nobody can own the Moon. Whether anyone can own what they dig out of it is the live argument.
• In practice, a commercial space lawyer spends far more time on FAA, FCC, ITU and export control paperwork than on treaty interpretation.

Ask ten people what space law is and most will describe something like maritime law for rockets. That is not far off, but it misses the part that matters commercially. Space law is not one rulebook. It is two layers that behave very differently.

The top layer is international. Five UN treaties negotiated between 1967 and 1979, plus a set of General Assembly principles and guidelines that are not binding but get followed anyway. This layer moves slowly. The last treaty was opened for signature in 1979 and almost nobody joined it.

The bottom layer is national. Every country that lets private operators launch or run satellites has built its own licensing regime, and those regimes are rewritten every few years. This is the layer that decides whether a company can fly.

If you only read the treaties, you will understand the constitutional theory and none of the practice.

The five treaties, and why only three matter day to day

The UN Office for Outer Space Affairs publishes the ratification tables every year. As at 1 January 2026 they looked like this.

Look at the last row. Seventeen parties, none of them a major spacefaring state. The Moon Agreement is the treaty everyone cites in arguments about lunar mining and the one nobody with a rocket has signed. It is a useful piece of evidence about what states were willing to accept in 1979, and very little else.

The Outer Space Treaty is the one that does the structural work. The Liability Convention and the Registration Convention fill in the detail on damage claims and on who has to tell the UN what they launched.

The sentence that built the whole licensing industry

Article VI of the Outer Space Treaty says that states bear international responsibility for national activities in outer space, including activities carried on by non-governmental entities, and that those activities require authorisation and continuing supervision by the appropriate state.

Activities of non-governmental entities in outer space, including the Moon and other celestial bodies, shall require authorization and continuing supervision by the appropriate State Party to the Treaty.

Outer Space Treaty, Article VI

Read that again with a commercial eye. A private company cannot be a party to the treaty. So the treaty reaches private activity by making the state answerable for it. If a US company puts a satellite in orbit and it causes damage, the claim under international law runs against the United States, not the company.

No government accepts that exposure without control. That is why the FAA licenses launches, the FCC licenses spectrum and orbital debris plans, the UK Civil Aviation Authority licenses UK operators, and India routes everything through IN-SPACe. Article VI is the reason those agencies exist in their current form.

Liability: two regimes, one of which has never been used

The Liability Convention splits damage into two categories, and the split matters enormously.

Under Article II, a launching state is absolutely liable for damage caused by its space object on the surface of the Earth or to aircraft in flight. No fault needs to be shown. If a stage falls on a house, the launching state pays.

Under Article III, damage caused anywhere other than the surface of the Earth, which in practice means collisions in orbit, is governed by fault liability. The claimant has to prove the other state was at fault.

The Convention then does something remarkable. It never says what the standard of care is. There is no definition of fault, no orbital rules of the road, no priority scheme. That gap is the single biggest reason the Liability Convention has never been formally invoked for an on-orbit collision, including after the 2009 Iridium and Cosmos collision that produced thousands of trackable fragments.

We wrote about how that plays out in practice in our piece on debris liability.

Nobody owns the Moon. Resources are a different question.

Article II of the Outer Space Treaty is one sentence long and it is the most argued over sentence in the field. Outer space, including the Moon and other celestial bodies, is not subject to national appropriation by claim of sovereignty, by means of use or occupation, or by any other means.

That settles territory. It does not obviously settle extracted material. Four states have now legislated to say that a company can own what it extracts: the United States in 2015, Luxembourg in 2017, the UAE in 2019 and Japan in 2021. Seventy states have signed the Artemis Accords, which record the view that extraction does not by itself amount to national appropriation.

Critics answer that Article VI attributes private conduct to the state, so a state licensing extraction is doing indirectly what Article II forbids directly. That argument is not going away, and the COPUOS working group on space resources is due to produce draft principles during 2026.

What a space lawyer actually does

Here is the honest version of the job, for anyone deciding whether to go into it.

Licensing and regulatory

Getting a client through the FAA’s Part 450 launch process, an FCC space station application, an ITU filing through a national administration, or the equivalent in the UK, India, Japan, New Zealand or Australia. This is the bulk of the work and it is closer to aviation and telecoms regulatory practice than to public international law.

Contracts

Launch services agreements, satellite procurement, hosted payload deals, ground segment, data licensing, in-orbit servicing. The distinctive feature is the cross waiver of liability, a reciprocal no-fault waiver that turns the whole supply chain into a set of self-insured silos. We explain how that works in our piece on launch contracts.

Export control

Spacecraft and their components sit on the US Munitions List under ITAR Category XV or on the Commerce Control List in the 9×515 series. Getting this wrong is a criminal exposure, not a commercial one. It occupies a lot of junior time.

Finance and insurance

Launch and in-orbit insurance placement, export credit, satellite backed lending, and increasingly the diligence around megaconstellation financing.

Public international law

Real but small. Treaty interpretation, COPUOS delegation work, advising governments on national legislation, academic writing. If this is the part you find interesting, plan to combine it with something that pays.

Where to start reading

Start with the Outer Space Treaty itself. It is about 2,500 words and you can read it in fifteen minutes. Then read Part 450 or the FCC’s Part 25, not because you will enjoy it, but because it will show you the distance between the constitutional layer and the working layer.

Our directory of regulators and primary sources links straight to the official text of everything mentioned here.