New countries should rent land—not buy it

Secession, war, and moon colonies aren't our only way to get new countries.

This is a guest essay by Julien Starr for Post Nation, seven writers exploring a world after nation-states. Support the project by collecting the series as a digital or print pamphlet. 👇🏻

Have you ever seriously thought about starting a new country? Not a thought experiment, not a flag and a website, but the real thing: a functioning state. What problem would it solve? What would its value proposition be? Who would want to live there, and why would any existing government help bring it into existence? Those are the questions that serious business founders ask, and there is every reason to apply the same discipline to statecraft.

Very few people alive today have had any hand in the creation of a new country. It is the ultimate startup and, by most reckonings, the ultimate challenge. The international order currently comprises 193 UN member states, the overwhelming majority of which came into existence through war, colonial withdrawal, imperial disintegration, or some combustible combination of the three. New countries have tended to be accidents of violence rather than products of intention. The Startup States model coined in 2025 challenges that assumption directly: a new country can be founded on dry land, by deliberate design, in full conformity with existing international law, without displacement of populations, without secession, and without any of the coercive machinery that has historically attended state formation.

The legal architecture to do this already exists. It has simply not been used in this way before.

The place to begin is with why the other routes do not get you all the way there, and why, given that you will have to negotiate with a government in any case, you should train for the gold medal rather than settle for something considerably less valuable.

There is a negotiating logic that runs through the entire Startup States approach. If you are going to invest the time, legal work, diplomatic effort, and relationship capital required to sit across a table from a sovereign government and make an ask, you may as well ask for the right thing. Think of it like the Olympics. Nobody trains for a bronze medal. Serious athletes train for gold, prepare for gold, and design every aspect of their preparation around winning it. An athlete who privately decides that silver is the more realistic target is not being pragmatic; they are negotiating against themselves before the competition has begun. The same principle applies here. Do not sell yourself short, go for the gold and go for starting your own country!

We can’t yet build countries at sea or on the moon

The high seas have a long romantic history as the venue of last resort for those seeking to escape the constraints of the existing territorial order, but the legal position is clear: Article 89 of the United Nations Convention on the Law of the Sea (UNCLOS) states that no state may validly subject any part of the high seas to its sovereignty. Article 60(8) adds that artificial islands and installations on the high seas have no territorial sea of their own and do not affect the delimitation of any maritime zone. The Montevideo Convention on the Rights and Duties of States (1933), which sets out the customary criteria for statehood in international law (defined territory, permanent population, government, and capacity to enter into relations with other states), further underscores this point. A floating platform in international waters cannot satisfy the Montevideo Convention’s requirement of a defined territory.

Proponents sometimes argue that a state not party to UNCLOS might recognise a maritime polity regardless. This misunderstands how the international order works. The Stimson Doctrine, first articulated in Henry Stimson’s note of January 1932 in response to Japan’s seizure of Manchuria, was a U.S. policy statement rejecting recognition of territorial or legal gains achieved through force or in violation of international obligations. Although initially a political declaration rather than binding law, it later informed a broader customary principle of non-recognition, which is reflected in Article 41 of the International Law Commission’s Articles on Responsibility of States for Internationally Wrongful Acts (2001), which requires states not to recognise as lawful situations created by serious breaches of peremptory norms of international law and not to assist in maintaining them. With 169 parties, UNCLOS represents one of the most widely subscribed multilateral instruments in existence; its signatories would bring considerable diplomatic pressure to bear on any non-signatory that contemplated recognising a polity constituted in defiance of the regime.

The same Stimson Doctrine logic applies to Antarctica, where the Antarctic Treaty of 1959 freezes all existing claims and expressly prohibits new ones, and to outer space, where the Outer Space Treaty of 1967 establishes that no celestial body is subject to national appropriation. The 56 parties to the Antarctic Treaty and the 115 parties to the Outer Space Treaty would invoke identical non-recognition machinery against any polity founded in violation of those instruments. The absence of a formal claim to Marie Byrd Land does not mean Antarctica is open for business.

The concept of terra nullius is similarly unavailing in the cases of disputed territory that occasionally excite the imagination of the new-country community. Terra nullius is a doctrine in international law meaning “land belonging to no one,” historically used to describe territory that is genuinely unclaimed and not under the sovereignty of any state, think Cabo Verde in the mid-15th century before the arrival of the Portuguese. The ICJ’s Advisory Opinion on Western Sahara in 1975 confirmed that terra nullius requires genuinely sovereignless territory, a bar that disputed parcels like Gornja Siga on the Danube between Croatia and Serbia, or Bir Tawil between Egypt and Sudan, simply do not meet. The analogy is a crayon on a table: if one sibling insists it belongs to the other and the other insists the same, that does not mean a cousin can walk in and claim it as his own. He could say so; it would not make it true. Croatia has, in practice, exercised the capacity to expel individuals attempting to establish a presence in Gornja Siga which it states belongs to the Republic of Serbia and Serbia says it belongs to the Republic of Croatia. The parcel has owners; they merely disagree with each other about which of them it is.

Digital state formation is a genuinely interesting project, and the impulse to build governance structures from first principles using voluntary participation rather than coercion is one the Startup States approach shares entirely. But international law has not caught up. No UN member state has yet extended de jure diplomatic recognition to a network state or DAO, permitting it to establish an accredited embassy and exchange diplomatic credentials. Estonia’s e-residency programme and Palau’s digital residency initiative are significant innovations but operate within the sovereign structures of those states, not as substitutes for them. Albania has given an AI system a ministerial advisory role, and reports have circulated that Antigua and Barbuda extended some form of recognition to the Joseon Empire, a claimed successor to Korea’s historical Joseon dynasty, though the precise status and scope of any such recognition, and whether any diplomatic mission has been established in St John’s, remain unverified at the time of writing. These are interesting data points. None of them constitutes a new subject of international law. The Montevideo criteria remain the governing standard, and a digital presence satisfies none of its four requirements: a defined territory, a permanent population, an effective government, and the capacity to enter into relations with other states.

Anything less than independence can be taken away

Suppose a founding community negotiates part of the way: a special economic zone, a charter city, an autonomous region with substantial self-governing powers. The political ask is smaller and the runway to launch is shorter. But the structural problem with even the most sophisticated subnational arrangement is that it is ultimately subject to the political authority of the state in which it sits. The rug can always be pulled.

Hong Kong is the most instructive example. Under the Sino-British Joint Declaration of 1984 and the Basic Law promulgated in 1990, Hong Kong maintained its own immigration controls, customs regime, legal system, and currency. It was, by most economic measures, one of the most successful jurisdictions in the world. But its autonomy was extended by a central authority and remained conditional on that authority’s continued forbearance. Events from 2019 onwards demonstrated what happens when forbearance runs out. Even at its peak, Hong Kong did not have a full seat at the table in international affairs. It was a guest at someone else’s table.

Próspera, the Zone for Employment and Economic Development established on Roatán in Honduras, represents perhaps the most sophisticated subnational governance experiment of the current generation: genuine rule-of-law infrastructure, investor protections, meaningful self-governance. When the Honduran government changed in 2022, the incoming administration moved to dissolve the ZEDE framework. Subsequent arbitration and political evolution have produced some stabilisation, and the current administration is considerably less hostile. But the episode established something that cannot be unestablished: the optionality to pull the rug was exercised, and that permanently changes the risk calculus for investors and founders alike.

The Turks and Caicos Islands make the same point from a different angle. In 2009, the United Kingdom suspended the territory’s locally elected government entirely, placing it under direct rule from London. Whatever one’s view of the merits, the episode confirmed that even a long-established, economically sophisticated British Overseas Territory can lose its autonomous governance at Westminster’s discretion. Subnational autonomy is, at its most fundamental level, a licence that can be revoked.

The lesson is consistent: if you are going to negotiate with a sovereign government, train for the gold medal.

The "Startup States Approach: No hidden agenda, just agreement to create a new country

The Startup States model is structurally distinct from any subnational arrangement, and this distinction matters as much in political framing as in legal substance. There is no Trojan horse here, no incremental accumulation of de facto autonomy that a host government might later come to resent. Our ask is explicit from the very first conversation: a new, internationally recognised state, constituted by treaty, on uninhabited and non-strategic territory that the host state already owns but is not using productively. Everything is on the table, upfront. That directness is a feature.

A government considering a subnational arrangement has legitimate reasons to worry that it is making a concession that will expand in ways it cannot anticipate. The Startup States approach eliminates that concern: The new state is sovereign from day one and is not seeking to grow its autonomy incrementally. The constitutional and territorial relationship between the two states is defined in the founding treaty and does not change unless both parties agree.

There is also a sovereignty argument that works in the host state’s favour. The act of bringing a new sovereign state into existence is one of the most powerful demonstrations of sovereignty that an existing state can make. You know you have free speech; but the fullness of that right becomes most vivid when you actually exercise it, especially when you say something consequential. A UN member state that midwifes a new country is not diminishing its own sovereignty. It is flexing it in the most unambiguous way available. This is scaling sovereignty, not eroding it.

The appropriate template for the parent-state role is Montenegro rather than Kosovo. Montenegro’s independence in 2006, achieved through a referendum that Serbia and Montenegro accepted and followed by prompt and widespread international recognition, confirms that consensual, parent-state-approved statehood produces impeccable legal standing from the outset. Kosovo’s trajectory was messier: the ICJ’s Advisory Opinion of 2010 confirmed that Kosovo’s 2008 declaration of independence from the Republic of Serbia which still contests Kosovo’s independence to this day did not violate international law, but recognition remains contested by a significant number of UN member states. For a Startup State whose entire value depends on unimpeachable legal standing, recognition-market fit is the key metric, and the Montenegro model delivers it. Moving fast and breaking things works for a consumer application; it is considerably less suitable when the thing you might break is your own legal basis for existence.

A critical structural concept here is the distinction between dominium and imperium: between ownership of territory, which remains with the host state throughout the life of the founding treaty, and the exercise of governmental jurisdiction over it, which vests in the new state from the moment the treaty enters into force. The host state never loses its land because it never loses legal title to the land it leases and it never loses ownership. It leases plenary jurisdiction over that land to the new state in exchange for among other indemnities rent payments. The host state becomes a landlord and even depending on the terms of the deal could become a dividend receiving shareholder in companies which may be developing the infrastructure and administering the services for the new country. There is no cession and therefore no secession, because territorial title never changes hands. The construction is analytically distinct from secession for five reasons: there is no unilateral act; no existing demos asserting self-determination; no impairment of the host state’s territorial integrity, because the arrangement is an exercise of that sovereignty rather than a violation of it; no violence or coercion; and no breach of any peremptory norm of international law.

The legal architecture

The legal instrument at the heart of the model is a founding treaty concluded under the Vienna Convention on the Law of Treaties of 1969, binding the parties under the principle of pacta sunt servanda and interpreted according to the Convention’s good faith rules. The treaty should be registered with the United Nations Secretariat pursuant to Article 102 of the UN Charter, both as a legal obligation of the host state and as a matter of strategic transparency, placing the new state’s foundational instrument on the permanent public record of the international community.

A common misconception is that Article 102 registration is available only to UN member states. It is not. The Cook Islands, a full subject of international law and a party to UNCLOS and numerous other multilateral instruments, is not a UN member but participates in treaty arrangements deposited or registered with the UN. A founding treaty concluded with a UN member state as co-signatory would be registrable under Article 102, providing formal documentary evidence of the arrangement’s consensual and transparent character.

The historical precedents for treaty-created states are more numerous than is commonly appreciated. The Lateran Treaty of 1929 between the Holy See and Italy constituted Vatican City as an independent sovereign entity upon a precisely defined parcel of territory within Rome, on a legal blank slate, by agreement between two willing parties: the single most directly apposite precedent for the Startup States construction in the history of international law. Singapore’s separation from Malaysia in 1965, effected by a Separation Agreement signed and ratified by both parties on the same day, resulted in UN admission six weeks later, confirming that consensual treaty-based state formation produces immediate and comprehensive international recognition. The Austrian State Treaty of 1955, concluding Austria’s post-occupation statehood between Austria and the four occupying powers, is a further illustration of the treaty as the foundational instrument of a new state’s legal personality.

The founding community will need to satisfy both the declaratory and constitutive theories of statehood. The declaratory theory, reflected in the Montevideo Convention, holds that statehood attaches upon satisfying the four criteria: permanent population, defined territory, effective government, and capacity for international relations. The constitutive theory holds that recognition by other states is itself a necessary element. A founding treaty with a recognising UN member state as co-signatory addresses the constitutive requirement directly. The concept of the inchoate state, an entity that satisfies some but not all Montevideo criteria at formation and may be treated as a state in statu nascendi during an initial developmental period, can be managed through transitional provisions in the founding treaty. The territorial dimension will need to engage with the principle of uti possidetis juris, confirmed by the ICJ in the Frontier Dispute between Burkina Faso and Mali in 1986, which provides the legal basis for the new state’s insistence on the stability of its founding boundaries.

What the host state gets: found money and a sovereignty flex

From the host state’s perspective, the proposition is at its most basic a real estate reclassification play with extraordinary upside. An uninhabited, non-strategic parcel of territory is a non-performing asset: it generates no revenue, may impose maintenance costs, and holds no diplomatic value in its current form. The founding treaty converts that parcel into an income-generating relationship while the host state retains full ownership and title throughout.

The economic logic of the reclassification is best illustrated by the property market on the French Riviera. Cross from the commune of Beausoleil into Monaco and property prices rise sharply, by multiples that no improvement in physical infrastructure could explain. The sun is identical. The sea is identical. The architecture is largely identical. What changes is the legal status of the territory, and that difference generates economic value placing Monaco among the highest per-square-metre jurisdictions on earth. Reclassifying idle land as a sovereign state is the ultimate rezoning decision. The land does not move. The title does not change. The legal status transforms, and so does the return on the asset.

Because ownership never changes hands, the arrangement is the structural opposite of colonialism: the financial flows run to the rightful owners of the land throughout the lease period. The host state receives rent and associated financial arrangements, converts a dormant asset into a productive one, and acquires a bilateral relationship with a state whose existence it made possible. This is found money, and there is no political embarrassment attached to it, because the host state never gave away the land. It rezoned it.

Where the host state has concerns about its governance authority, a condominium structure, under which both states exercise defined concurrent functions over the same territory, is a calibrated and well-precedented solution. International law has long recognised concurrent sovereignty as a legitimate arrangement, as the historical examples of the Anglo-Egyptian Sudan under the Agreement of 1899 and the New Hebrides under the Convention of 1906 demonstrate. Under such a condominium, the host state retains defined governance levers agreed in the founding treaty while the new state exercises independent legal personality in foreign affairs, immigration, and customs. Sovereignty is not lost; it is shared and, in sharing, amplified.

The Startup States approach requires no amendment to the UN Charter, no revision of the Vienna Convention, and no departure from the Montevideo criteria. It requires a willing host state, a serious founding community, and the conviction to deploy treaty law in a way that the law already fully permits. Vatican City proves the model is not theoretical. Singapore proves the recognition follows swiftly when the consent is genuine.

What the model delivers in return is the gold medal: a fully sovereign state, constituted by treaty, registered with the United Nations, entitled to all the privileges and protections of membership in the international community, with its own foreign policy, its own immigration and customs regime, and a standing in international affairs that cannot be revoked by a change of government in someone else’s capital. Not a subnational arrangement that a legislative session can unwind. Not a digital jurisdiction awaiting recognition that has not yet come. A state whose founding instrument is public, whose legal basis is unimpeachable, and whose sovereignty is not borrowed from a central authority.

The cap table is clean from day one. The legal framework is already in place. Train for the gold. Go build a country.