Daniel sent us this one — he's asking about the weird world of nested leases, which is one of those topics that sounds dry until you realize it determines whether you actually own your home or just own the right to worry about someone else's expiration date. He's pointing to Israel's bizarre arrangement where the state leases land from churches, then subleases to developers, who sell apartments to buyers who think they own something — and he's asking how deep this recursion can go, and how today's lease structures stack up against historical land ownership. Honestly, this is the kind of question that keeps me up at night. Or would, if I weren't a sloth.
Who actually owns the dirt? To answer that, we need to understand what a leasehold actually is — and it's not what most people think. Most people assume you either own your home or you rent it. Leasehold sits in this strange middle ground. You own a time-limited interest in the property — you can sell it, mortgage it, pass it to your heirs — but you don't own the land underneath. That belongs to someone else, and when the clock runs out, the whole thing reverts to them. The legal term for that snapback is "reversion." And the person who owns the reversion holds what's called the freehold, or in some systems the head lease.
It's ownership with an expiry date baked in. Like buying a carton of milk, except the milk costs half a million dollars and you're hoping the dairy doesn't change the terms when you need a refill.
That's actually not a bad way to put it. And the key distinction here is between a ground lease and an occupancy lease. A ground lease is typically long-term — forty-nine years, ninety-nine years, sometimes nine hundred ninety-nine years — and it covers the land itself. The person who holds the ground lease then builds on it or develops it, and they might sell or sublease individual units. Those subleases are occupancy leases. So you get layers. The freeholder leases the land to a developer, the developer builds a building and subleases individual apartments to homeowners. That's two layers right there. In Jerusalem, it goes deeper.
Let's start with the most extreme case of leasehold nesting in the world today: Israel. And within Israel, Jerusalem is the epicenter. So walk me through the layers, Herman. Who owns what?
Layer one — the state. The Israel Land Authority, the ILA, owns ninety-three percent of the land in Israel. That's an extraordinary number. By comparison, the UK government owns about twelve percent of land. The US federal government owns about twenty-eight percent. Israel's ninety-three percent puts it in a category with Singapore and a few other places where the state is basically the landlord of the entire country.
The OPEC of dirt, as we've discussed before.
The ILA was established in nineteen sixty, and it doesn't sell land. It leases it. Residential leases are typically forty-nine or ninety-eight years. Commercial leases can be shorter. And the renewal fee — and this is where it gets spicy — the renewal fee is set at ninety-one percent of the land's value at the time of renewal. So you've paid off your mortgage, you've lived in your home for two generations, and then the state says, "That'll be another ninety-one percent of the land value, please.
Which is effectively a second purchase price, minus a nine percent courtesy discount. Very generous of them.
The nine percent is almost insulting, honestly. It's like they're saying, "We could take it all, but we're leaving you just enough to feel like you negotiated something.
That's just the state layer. In Jerusalem, it gets weirder.
Layer two — the churches. The Greek Orthodox Patriarchate of Jerusalem owns approximately twenty percent of the land in the Old City and significant chunks of West Jerusalem. We're talking about some of the most valuable real estate on the planet. Rehavia, Talbiyeh, parts of the city center — all built on church land. And here's where the nesting begins. In the nineteenth century, the Ottoman Empire granted the Patriarchate what were essentially perpetual leases over these lands. The Ottomans had this system called the miri system where the state technically owned agricultural land but granted usufruct rights — the right to use and profit from the land — to individuals or institutions. The churches got these grants, and they held onto them.
The Ottoman Empire is gone, but the lease isn't.
The lease survived the empire. That's the first thing people don't understand. When the British took over after World War One, they inherited the Ottoman land registry. When Israel was established in nineteen forty-eight, it inherited the British registry. The legal chain was never broken. So the Greek Orthodox Church holds a lease — or something very close to a lease — that dates back to Ottoman rule. Then in the nineteen fifties and sixties, the Israeli state expropriated much of this church land under various laws and then leased it to developers. Those developers built apartment buildings and sold ninety-nine-year leasehold interests to individual buyers.
A homeowner in Rehavia today is holding a sub-sub-lease. The church has the Ottoman grant, the state has a lease from the church, the developer had a lease from the state, and the homeowner has a lease from the developer. That's four layers.
At least four. And in some cases, there might be intermediate holding companies or management entities that add a fifth layer. The question the prompt asks is whether there's a limit to this recursion. How deep can it go before the whole thing becomes legally meaningless?
Before we get to the limit, what happens when a lease at the top expires? If the church's Ottoman lease has some kind of end date — or if the state's lease from the church expires — does the whole chain collapse? Does the homeowner's lease just vanish?
This is the nightmare scenario, and it's not theoretical. In English common law, which heavily influenced Israeli property law, when a head lease expires, all underleases derived from it also expire. The legal principle is nemo dat quod non habet — no one can give what they don't have. If the state's lease from the church ends, the state can no longer grant occupancy rights, and the homeowner's interest evaporates. Now, in practice, courts and legislatures usually step in to prevent mass dispossession, but the legal vulnerability is real.
We're seeing this play out right now. The ninety-nine-year leases on church land in Jerusalem — many of them signed in the nineteen twenties and thirties — are expiring between twenty twenty-six and twenty thirty-five. Thousands of homeowners are sitting on properties where the clock is running out.
And the renewal terms are being negotiated right now, as we speak. The Knesset has been debating lease reform bills. There was a bill in twenty twenty-four that proposed mandatory lease-to-freehold conversion for residential properties — basically forcing the ILA to sell the land to homeowners instead of renewing the lease. It failed in committee, but the pressure is building. Meanwhile, the churches are demanding renewal premiums of fifty to eighty percent of the property's current market value. For a family that bought an apartment for, say, two million shekels thirty years ago, and it's now worth four million, the church might demand two to three million shekels to renew the lease for another ninety-nine years.
That's not a renewal fee. That's a ransom.
Unlike the ILA's renewal formula, which is at least codified in law — ninety-one percent of land value — the church leases have no standard formula. It's whatever the church can negotiate.
Let's talk about the recursion limit. You mentioned nemo dat — no one can give what they don't have. Is there a point where the legal chain gets so long that the original freeholder can't enforce anything against the person at the bottom?
There are two concepts in property law that create practical limits. The first is privity of estate and privity of contract. Privity of contract means you can only enforce terms against the person you signed a contract with. If I lease land to you, and you sublease to someone else, I can enforce the head lease against you, but I have no direct contractual relationship with the subtenant. Privity of estate gives me some rights against the subtenant — like collecting rent or enforcing covenants that run with the land — but it's weaker, and it gets weaker with each layer. By layer four or five, the original freeholder's ability to enforce anything is mostly theoretical.
At a certain depth, the leash breaks. The freeholder can pull, but nobody at the bottom feels it.
And this was a real problem in medieval England, which I want to get to in a minute. The second limit is the rule against perpetuities, which exists in some form in most common law jurisdictions. The basic idea is that you can't create property interests that vest too far in the future. In its classic formulation, an interest must vest within twenty-one years after the death of a person alive at the time the interest was created. That puts a soft cap on how long leasehold chains can extend.
Israel doesn't use the rule against perpetuities in the same way. The Land Law of nineteen sixty-nine caps residential leases at ninety-eight years. So Israel limits recursion by time rather than by depth. You can have as many layers as you want, but they all have to fit within ninety-eight years.
Which is a much cruder tool, honestly. The rule against perpetuities is elegant — it says "no dead-hand control beyond a reasonable horizon." Israel's approach says "everything expires in ninety-eight years, good luck." It doesn't prevent deep nesting; it just guarantees that the whole nesting doll explodes at a fixed date.
There was a case that illustrated exactly this problem. Kfar Shmaryahu, twenty twenty-three. What happened there?
Kfar Shmaryahu is an affluent community north of Tel Aviv, built on church land. The ILA held the master lease from the church, and when the master lease came up for renewal, the ILA and the church couldn't agree on terms. The ILA essentially refused to renew. For eighteen months, about three hundred homeowners were in legal limbo. They couldn't sell their homes because no buyer would touch a property with an expired head lease. They couldn't get mortgages. Their property values effectively went to zero overnight.
These are people who thought they owned their homes. They had deeds, they had mortgages, they had been paying property taxes for decades. And suddenly they discover they're tenants in a dispute between two giant landlords that don't even know their names.
The case was eventually resolved through political intervention — the government leaned on the ILA to renew — but it exposed the fragility of the whole system. Every homeowner in Israel on church land is one lease negotiation away from the same nightmare.
There's another famous case — the Mamilla Cemetery controversy. That's a different flavor of the same problem.
Mamilla is adjacent to the Old City. It's a historic Muslim cemetery, managed by the Islamic Waqf. In the nineteen sixties, the state expropriated part of the land and leased it to developers. Today there's a shopping center, a parking lot, and a luxury hotel sitting on that land. The Waqf has challenged the leases in court, arguing the expropriation was illegal and the original lease terms were violated. The case has been bouncing through Israeli courts and international forums for decades. It's a parking lot built on a lease dispute built on an expropriation built on an Ottoman-era religious endowment.
We've seen how deep the recursion goes in practice. But is this really new? Let's go back eight hundred years to when recursion was even deeper.
This is where it gets fascinating. In feudal England, the king technically owned all the land. Everyone else held land as a tenant of the king, directly or indirectly. The king would grant a fief to a tenant-in-chief — a baron or a bishop. That baron would subinfeudate, granting portions of his fief to sub-tenants, who would grant to their own sub-tenants, and so on. By the thirteenth century, it was common to have five to seven layers of tenancy between the king and the actual person farming the land.
And we think four is bad.
And each layer came with feudal obligations — military service, agricultural labor, payments in kind. The problem was that by layer six or seven, nobody could figure out who owed what to whom. The king couldn't collect services from someone seven links down the chain. The whole system became administratively unmanageable.
They fixed it. Statute Quia Emptores, twelve ninety.
Quia Emptores — Latin for "since the purchasers" — was passed under Edward the First. It banned subinfeudation outright. From that point on, if a tenant sold his land, the buyer didn't become a sub-tenant — the buyer stepped into the seller's shoes and held directly from the seller's lord. No new layers. It froze the recursion depth where it was and gradually simplified the system over the following centuries.
The recursion limit was discovered empirically. The system broke at about layer seven, and the law stepped in to cap it. Modern leaseholds rarely exceed three layers, not because we're smarter, but because we've inherited legal infrastructure that prevents the old problems from recurring.
Because modern property registration systems impose transparency costs. Every layer of lease has to be registered, recorded, and searchable. In Israel, the Tabu — the land registry — is still paper-based in many districts. You want to trace the chain of leases on a Jerusalem apartment? You might be digging through Ottoman-era documents in a basement office. The opacity itself acts as a brake on recursion because at a certain point, nobody can prove their rights well enough to enforce them.
The modern problem isn't that recursion is too deep — it's that it's too opaque. Feudal recursion was deep but relatively transparent within the community. Everyone knew who the lord was, who the sub-lord was, who farmed which strip. Modern recursion is shallow — two or three layers — but the layers are corporate entities, holding companies, offshore trusts. You can't walk down to the village square and ask who actually owns the dirt.
That opacity has real economic consequences. A twenty twenty-two Bank of Israel study found that leasehold properties sell at a fifteen to thirty percent discount compared to equivalent freehold properties. That's not a small haircut. On a four million shekel apartment, that's six hundred thousand to one point two million shekels of value that simply doesn't exist because the ownership structure is uncertain.
It's not just Israel. Hong Kong is the poster child for leasehold uncertainty. All land in Hong Kong is owned by the state — China since nineteen ninety-seven, Britain before that. Residential leases are typically fifty years, and a huge number of them expire in twenty forty-seven. That's about two and a half million residential units. The market is already pricing in a "leasehold risk premium" of twelve to eighteen percent. People are buying apartments knowing that in twenty-one years, the legal basis for their ownership might evaporate.
Twenty forty-seven is the big one because that's when the fifty-year clock from the handover runs out. The Sino-British Joint Declaration guaranteed fifty years of continuity, but it says nothing about what happens after. China has been vague. "We'll figure it out." And the Hong Kong property market — one of the most expensive in the world — is built on that vagueness.
The UK has its own leasehold scandal, which is less about geopolitics and more about predatory contract design. Ground rents doubling every ten years, homes becoming unsellable.
The UK situation is a masterclass in how leaseholds go wrong at scale. Developers sold new-build houses as leasehold rather than freehold, then sold the ground rents to investment funds. The ground rents had escalation clauses — doubling every ten years was common. So a ground rent of two hundred fifty pounds a year becomes five hundred, then a thousand, then two thousand. Within thirty years, the ground rent exceeds the mortgage payment. Buyers couldn't sell because no lender would touch a property with an escalating ground rent that could theoretically reach tens of thousands of pounds.
The house is otherwise a normal house. It's not a flat in a building with common areas that need maintenance. It's a detached house with a front garden and a driveway, and someone else owns the dirt under it and charges you for the privilege of standing on it.
The Leasehold Reform Act of twenty twenty-two banned ground rents on new leasehold houses, but it didn't touch existing leases, and it didn't cover flats. There are four point three million leasehold flats in the UK that are completely unaffected by the reform. The scandal is very much ongoing.
Let's talk about the contrast with systems that solved this problem differently. The US has almost no leasehold residential property outside of a few co-op buildings in New York and, interestingly, Hawaii.
Hawaii is the American exception that proves the rule. Much of the land in Hawaii was historically owned by a small number of trusts and estates — the Bishop Estate, the Campbell Estate — descendants of the old plantation families. They leased land to homeowners on long-term ground leases. In the nineteen sixties and seventies, as those leases started expiring, homeowners faced the same renewal shock we're seeing in Jerusalem. The Hawaii state legislature eventually passed lease-to-fee conversion laws that forced landowners to sell the fee simple — the freehold — to the leaseholders at a formula price. It was messy, it was litigated for decades, but it worked. Most leasehold residential property in Hawaii has now been converted to freehold.
Hawaii had its Kfar Shmaryahu moment, but at state scale, and they legislated their way out of it. Israel's twenty twenty-four bill tried to do the same thing and failed.
The politics are different. In Hawaii, the leaseholders were local voters, and the landowners were trusts perceived as colonial-era holdovers. In Israel, the landowner is the state itself, and converting leaseholds to freeholds would mean the state giving up control over ninety-three percent of the land. That's not just a property reform — it's a fundamental redefinition of the relationship between the state and the land. The ILA and its predecessors were created specifically to hold land in perpetuity for the Jewish people. Selling it off, even to Israeli citizens, is ideologically fraught.
There's also the German system, the Erbbaurecht, which is a ninety-nine-year ground lease with automatic renewal rights. It's designed to avoid the renewal shock problem entirely. When the lease expires, it renews automatically unless the freeholder can show a specific reason to terminate, and the renewal terms are regulated.
The Erbbaurecht is a genuinely elegant solution. It separates land ownership from building ownership permanently, but it gives the building owner security. You know the lease will renew. You know the formula. You can plan your life around it. It's used extensively for single-family homes and for cooperative housing projects. About five percent of German residential property is on Erbbaurecht land, and it works because the rules are clear and the renewal is automatic.
Contrast that with Jerusalem, where the renewal terms on church land are whatever the church can extract in a one-on-one negotiation with a terrified homeowner who has no leverage.
The homeowner's only leverage is political — making enough noise that the Knesset steps in. Which is a terrible way to design a property system.
Alright, let's address some misconceptions before we get to the practical takeaways. First one: leasehold is the same as renting. This is wrong, and it's important to understand why.
A leasehold is a transferable property interest. You can sell it, mortgage it, bequeath it, use it as collateral. A rental agreement — a tenancy — is a personal contract. You can't sell your rental lease. You can't pass it to your kids. The distinction is legally fundamental. A ninety-nine-year leasehold is ownership of a time-limited right, and for most practical purposes, during those ninety-nine years, it functions like ownership. You can renovate, you can build, you can sublease — subject to the terms of the head lease, which is where the trap lies.
Second misconception: Israel's state ownership of ninety-three percent of land is unique. It's not. Singapore is at about eighty percent state-owned. Canada has eighty-nine percent Crown land. Sweden is around seventy percent. What's unique about Israel is the nesting — the church leases within state leases — and the absence of a right to buy the freehold.
In the UK, leaseholders have enfranchisement rights — the legal right to force the freeholder to sell them the freehold at a fair market price. It's not automatic, and it's not cheap, but it exists. In Israel, there is no enfranchisement right. The ILA does not sell land. There are a few exceptions — some agricultural communities, some industrial zones — but for residential property, you cannot force the state to sell you the dirt under your house.
Third misconception: leasehold recursion is a modern problem created by complex financial instruments. In reality, feudal systems had deeper recursion and solved it centuries ago. Modern recursion is shallower but more opaque. The problem isn't depth — it's transparency.
The feudal solution — Quia Emptores — was to ban new layers. The modern solution might be the opposite: make layers fully transparent so that depth doesn't matter. Estonia is piloting a blockchain-based land registry that tracks the full chain of title, including leasehold interests. Every layer is visible, every covenant is searchable, every expiry date is flagged. If that system were applied to Jerusalem's church land, a homeowner could see at a glance: "My head lease expires in twenty thirty-one. My renewal formula is unregulated. My risk exposure is high." That transparency alone would transform the market.
All this history and legal complexity boils down to one question for anyone buying a home: how much of your property do you actually control? Here's how to find out.
This is where we get practical. If you own or are considering a leasehold property — whether in Israel, the UK, Hong Kong, or anywhere else — there are three things you need to check. First, the head lease expiry date. Not your lease — the master lease at the top of the chain. If the head lease expires in fifteen years and your sublease has thirty years left, your sublease is worth zero after year fifteen. The head lease expiry is the hard ceiling.
Second, the renewal formula. Is it codified in law, like the ILA's ninety-one percent of land value? Is it negotiated case by case, like the Jerusalem church leases? Is it automatic, like the German Erbbaurecht? If there's no formula, you're gambling. And the house always wins.
Third, calculate what I call the "effective ownership fraction." This is a back-of-the-envelope way to quantify your actual control. Take the remaining years on the shortest lease in the chain — usually the head lease. Divide by ninety-nine, which is the standard for a "full" leasehold. If you have fifty years left, your effective ownership fraction is about fifty percent. Below thirty years, it drops below thirty percent. Below twenty years, you effectively own nothing — you're a tenant with a very long eviction notice.
This maps to market reality. Properties with less than fifty years on the lease trade at steep discounts. In the UK, lenders won't even issue mortgages on properties with less than seventy years remaining. In Hong Kong, the discount starts appearing at about thirty years from the twenty forty-seven cliff.
For investors, the actionable insight is: never buy a leasehold property without understanding the head lease expiry and the renewal formula. If the formula is unregulated — as it is on Jerusalem church land — negotiate a cap before you close. Get it in writing from the freeholder. If they won't give you a cap, walk away. The discount you're getting on the purchase price is not a bargain — it's compensation for a risk you can't quantify.
For policymakers, the actionable insight is that the recursion limit is not a legal problem — it's a transparency problem. Israel's Tabu land registry is still paper-based in many districts. You can't digitize what you can't find. A blockchain-based registry that tracks the full chain of leases, as Estonia is piloting, would reduce uncertainty and unlock billions in property value without changing a single law.
For homeowners in Jerusalem sitting on church land with leases expiring in the next decade — organize. The Kfar Shmaryahu homeowners got their resolution through collective action and political pressure. Individual negotiations with the church will not go well. Collective bargaining, possibly through a homeowners' association that spans multiple buildings on the same church parcel, changes the power dynamic entirely.
Leaseholds are ancient, recursive, and surprisingly resilient. But as we look to the future, one question keeps nagging at me. As climate change forces mass migration, are we going to see a return to recursive leaseholds as a way to manage land use rights in refugee-hosting regions?
It's already happening. Jordan and Lebanon use ninety-nine-year leases for Syrian refugee settlements. The host country retains sovereignty over the land, the UN or an NGO holds a head lease, and the refugees hold occupancy rights. It's a three-layer leasehold system, created ad hoc, with no clear renewal mechanism.
It's Jerusalem church land, but with millions of people and humanitarian stakes instead of property values.
The other frontier is digital land. Decentraland, The Sandbox — these are virtual worlds where you "buy" land. But what are you actually buying? A license from the platform owner. A time-limited right to use a digital parcel. The platform can change the terms, revoke access, or shut down entirely. That's not ownership. That's a leasehold from a feudal lord who didn't even bother to write Quia Emptores.
We're recreating thirteenth-century England in the metaverse. Seven layers of tenancy, no transparency, and the king can revoke your fief with a terms-of-service update.
The lo-fi girl is the George Owen Squier of the twenty-first century, and the metaverse landlord is Edward the First, but with worse graphics.
On that note — Hilbert, save us from ourselves.
Now: Hilbert's daily fun fact.
Hilbert: The word "steelyard" — as in the Hanseatic League's London trading post — shares a root with the Old Norse "stál" meaning "a place for weighing goods." The Hanseatic merchants were required by their own trade rules to weigh all goods on the Steelyard's official scales, and the scale itself — a balance with an adjustable counterweight — took its name from the yard where the weighing happened. By the nineteen seventies, the last remaining Hanseatic steelyard scale in active commercial use was at a fish-packing plant in Resolute, Nunavut, where it had been brought by a Danish Greenland trader in nineteen fifty-two and was still used to weigh char and turbot until the plant switched to electronic scales in nineteen seventy-eight.
...right.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop for that voyage to Resolute by way of medieval fish scales. If this episode made you think differently about the ground beneath your feet — or made you want to check your head lease expiry date in a cold sweat — leave us a review and tell us your weirdest leasehold story.
We're at myweirdprompts dot com. Until next time.
