The Registration Convention Explained: Who Registers a Satellite and Why It Decides Who Controls It

Every satellite launched into orbit is supposed to belong, legally, to a State. Not in the sense of ownership, which can sit with a company in any country, but in the sense of jurisdiction: one State whose law governs the object and whose registry it appears on. The instrument that makes that link is the Convention on Registration of Objects Launched into Outer Space, usually called the Registration Convention.

It is the least discussed of the five UN space treaties, and probably the most administrative. It is also the one that quietly decides which country’s courts, regulators and criminal law follow a satellite into orbit. With tens of thousands of objects now tracked and private companies buying, selling and servicing satellites in space, the gaps in a 1970s registration scheme have started to matter. This explainer sets out what the convention requires, who has to register, and where it runs out.

What the Registration Convention is

The convention was negotiated in the Legal Subcommittee of the UN Committee on the Peaceful Uses of Outer Space. The General Assembly commended it by resolution 3235 (XXIX) on 12 November 1974. It opened for signature in New York on 14 January 1975 and entered into force on 15 September 1976. The UN Secretary-General is the depositary. The full text is on the UNOOSA website and runs to twelve articles.

It did not start from nothing. In December 1961 the General Assembly, in resolution 1721 B (XVI), had already asked States launching objects into orbit to provide information to the UN for a public register. The convention turned that voluntary practice into a treaty obligation and, as UNOOSA puts it, “expanded the scope” of the existing UN register. States that are not parties still submit information under resolution 1721 B, so the UN register contains objects registered under both routes.

According to the UN Treaty Collection, there are 78 parties as of 18 October 2026. Burundi, Iran and Singapore have signed but not ratified. Recent accessions include the Philippines (14 May 2024), Ecuador (17 March 2025) and Malaysia (26 September 2025). Four intergovernmental organisations have declared acceptance under Article VII: the European Space Agency (1979), EUMETSAT (1997), the European Telecommunications Satellite Organization (2014) and Intersputnik (2018). For comparison, UNOOSA’s status table as at 1 January 2025 recorded 116 parties to the Outer Space Treaty, so a substantial number of treaty parties still sit outside the registration regime.

Two registers, one national and one international

The convention creates a two-level system.

At the national level, Article II requires that “when a space object is launched into earth orbit or beyond, the launching State shall register the space object by means of an entry in an appropriate registry which it shall maintain.” The State must tell the Secretary-General that it has set up such a registry. Article II(3) leaves the contents of the national registry and the conditions under which it is kept to each State.

At the international level, Article III requires the Secretary-General to maintain a register, with “full and open access” to its contents. UNOOSA runs it in practice and publishes the submissions and an online index. One detail surprises many operators: UNOOSA says registration with the Secretary-General can only be made by the government of a State of registry through its accredited permanent mission to the UN, or by the headquarters of an organisation that has accepted the convention. Submissions sent directly by companies, universities or even national space agencies are not treated as valid.

What has to be registered

Article IV lists the information each State of registry must furnish “as soon as practicable”:

(a) the name of the launching State or States;

(b) an appropriate designator of the space object or its registration number;

(c) the date and territory or location of launch;

(d) basic orbital parameters, namely nodal period, inclination, apogee and perigee; and

(e) the general function of the space object.

States may add information “from time to time” and must notify the Secretary-General, “to the greatest extent feasible and as soon as practicable”, of objects that were but no longer are in Earth orbit. Article V adds that where an object is marked with its designator or registration number, the State should say so.

That is the whole of the mandatory content. There is no field for the owner, the operator, the frequencies used or the end of life plan. “General function” can be as vague as “communications”. And “as soon as practicable” sets no fixed period.

The General Assembly tried to raise the standard in resolution 62/101 of 17 December 2007, “Recommendations on enhancing the practice of States and international intergovernmental organizations in registering space objects”. It recommends using the COSPAR international designator, Coordinated Universal Time for launch dates and standard units for orbital parameters, and providing additional information such as geostationary orbit location, changes of operational status, the approximate date of decay or re-entry, and moves to a disposal orbit. It says expressly that nothing in it “constitutes an authoritative interpretation of or a proposed amendment to” the convention. UNOOSA now offers a model registration form built around these recommendations.

Who has to register: the launching State problem

The convention borrows its definition of “launching State” from the Liability Convention. A launching State is “a State which launches or procures the launching of a space object” or “a State from whose territory or facility a space object is launched”. “Space object” includes component parts and the launch vehicle.

For a single government satellite launched from its own territory, that is easy. Modern missions are rarely like that. A satellite built and operated by a company in one country, procured by a customer in a second and launched from a third may have three or more launching States. Article II(2) says that where there are two or more, “they shall jointly determine which one of them shall register the object”, bearing in mind Article VIII of the Outer Space Treaty and without prejudice to agreements between them on jurisdiction and control.

The convention does not say what happens if they do not. Resolution 62/101 recommends that the State from whose territory a launch takes place should, absent prior agreement, contact the other potential launching States to settle who registers, that objects in a joint launch should be registered separately, and that each should go on the registry of the State responsible for its operation under Article VI of the Outer Space Treaty. It also suggests that launch providers advise their customers to approach the appropriate State. All of this is good practice, not obligation.

The result is that registration often follows licensing. A State that authorises and supervises an operator’s activities is the natural State of registry, which is why the question of Article VI responsibility under the Outer Space Treaty and the question of registration tend to be answered together.

Why registration matters legally

The convention looks administrative, but three other treaty provisions give it teeth.

The first is jurisdiction and control. Article VIII of the Outer Space Treaty provides that a State “on whose registry an object launched into outer space is carried shall retain jurisdiction and control over such object, and over any personnel thereof, while in outer space or on a celestial body.” Registration is therefore the hook for applying national law in orbit, including criminal law, regulatory conditions and court jurisdiction over events on board. The same article confirms that ownership is “not affected” by presence in space, so registration and ownership are separate questions.

The second is identification after an accident. The convention’s preamble says it is meant to give States “additional means and procedures to assist in the identification of space objects”. Article VI goes further: where a State cannot identify an object that has caused damage, other parties, “in particular States possessing space monitoring and tracking facilities”, should help on request. Identification is the first step in any claim under the Liability Convention, which we explain in our piece on who pays when satellites collide. The 1978 crash of the nuclear powered Cosmos 954 in Canada, covered in our round-up of space law cases, showed how quickly identification becomes a legal question.

The third is return. Under Article VIII of the Outer Space Treaty, objects found beyond the limits of the State of registry “shall be returned to that State Party, which shall, upon request, furnish identifying data prior to their return”.

Registration also features in newer political instruments. Section 7 of the Artemis Accords commits partners in a joint mission to decide between themselves which of them should register a relevant space object, and to consult non-parties to the convention where they are involved.

How it works in practice: the UK example

National systems vary, but the UK shows how a State fits the convention into domestic law. Section 7 of the Outer Space Act 1986 requires the Secretary of State to “maintain a register of space objects”, including such particulars “as the Secretary of State considers appropriate to comply with the international obligations of the United Kingdom”, and to let the public view it free of charge. Section 61 of the Space Industry Act 2018 adds a public register of launches from UK spaceports.

The UK regulator, the Civil Aviation Authority, publishes two registers. The UK registry of space objects lists objects for which the UK is the launching State, in line with the convention. A separate UK supplementary registry lists objects for which UK licences have been issued but the UK is not the launching State, including objects where “it was jointly determined that another ‘launching State’ should register that object with the United Nations.” That second register is a practical answer to a problem the convention never addressed: a State can supervise an activity without being the State of registry.

The gaps

Four weaknesses are worth knowing.

There is no deadline. “As soon as practicable” lets registrations lag behind launches, sometimes by a long time, and nothing in the convention penalises delay or omission.

There is no rule for transfers in orbit. Satellites are now sold, leased and moved between operators while in space. If the buyer is in a State that is not a launching State, that State cannot become the State of registry under the convention’s own definitions, even though it may now be the one supervising the activity. Resolution 62/101 recommends that the State of registry, in cooperation with the State responsible under Article VI, could report the date of the change in supervision, the new owner or operator, and any change of orbital position or function. That is a reporting fix, not a transfer of jurisdiction.

The information is thin. The five mandatory items were designed for a few hundred government satellites, not for large constellations, servicing vehicles that dock with other spacecraft, or lunar landers that leave hardware on the surface. None of them tells another operator how to reach the people flying the satellite, which is the information that matters for collision avoidance and one reason space traffic management still has no referee.

And participation is incomplete. With 78 parties, a significant part of the world registers, if at all, under a 1961 resolution rather than a binding treaty.

What I would watch

The first is in-orbit transfers. As servicing missions and secondary markets for satellites grow, I expect more States to follow the UK in keeping supplementary registers, and more pressure to treat the change of supervision recommendations in resolution 62/101 as standard practice.

The second is the Moon. Landers and rovers are space objects, and joint lunar missions involve several agencies and contractors. The Artemis Accords’ registration clause will be tested by the first missions in which two or more partners each have a plausible claim to be the launching State.

The third is accessions. The Philippines, Ecuador and Malaysia have joined since 2024. Each new party adds a national registry and a binding duty to report. The States that launch the most already belong, so the gains now come from newer space nations that procure launches abroad.

The fourth is national licensing. Because the convention leaves the content of national registries to each State, the real improvements in transparency are likely to come from licence conditions rather than treaty amendment. Regulators can require operators to provide far more than the five items in Article IV as a condition of their licences.

Anuj KumarEditor-in-Chief

Anuj Kumar is the founder of Legal Desire Media and publishes Fashion Law Journal and Space Law Journal.

Space Law Journal is a joint initiative of Legal Desire Media and Odyogeek. Spotted something we got wrong, or want to respond? Write to space@legaldesire.com.