Ask someone who owns the most land and they will name the wrong person. A Saudi royal. Jeff Bezos. Roman Abramovich, or whoever replaced him in the cultural inventory of ostentatious ownership. The answer feels intuitive because it is produced by a sensible mental model: extreme wealth concentrates in the hands of visible individuals, and land, being the most tangible of assets, would naturally be among the things those individuals own.<\/p>\n\n\n\n
The answer is wrong. Not marginally wrong \u2014 wrong in a way that reveals something important about how economic power constructs its own invisibility. The world’s largest landowners are not on any wealth ranking. They do not file consolidated accounts. They do not appear in the annual surveys of billionaires that populate the financial press. They are an ancient Church whose property cannot be tallied because it is held across thousands of legal entities that predate modern accounting, a constitutional abstraction that belongs to no individual and answers to no shareholder, agricultural dynasties in Brazil whose holdings exceed the land area of entire European nations, and pension funds managing the retirement savings of schoolteachers who have no idea they are, in a meaningful sense, landlords. The category error is not a gap in the data. It is a structural feature of the ownership itself.<\/p>\n\n\n\n
The Catholic Church is estimated to hold somewhere in the range of 70 to 80 million hectares of land globally. The qualification is necessary and, as it turns out, analytically important: these figures \u2014 circulated by land-mapping researchers and widely repeated in outlets covering global resource ownership \u2014 do not derive from any primary accounting source, because no such source exists or could exist. The Church holds property through thousands of separate legal entities \u2014 dioceses, religious orders, pontifical universities, charitable foundations, national churches across 193 countries \u2014 none of which consolidate their balance sheets. The Vatican’s Secretariat for the Economy, established in 2014 under Cardinal George Pell as part of an effort at financial transparency, has struggled to account for even the directly Vatican-administered properties. The canonical concept of the “public juridic person” \u2014 the legal form through which most Church property is held \u2014 was not designed with transparency in mind.<\/p>\n\n\n\n
The Juridic Person Problem<\/strong>\nUnder canon law, a \"public juridic person\" is an entity \u2014 a diocese, a religious institute, a pontifical foundation \u2014 that acts in the name of the Church and holds property on its behalf. The Vatican Secretariat for the Economy, established by Pope Francis in 2014 and modelled partly on national finance ministries, has attempted to bring consolidated oversight to this system. The effort has been instructive precisely because of what it revealed: the 2019 London property scandal, in which the Vatican's Secretariat of State was found to have invested funds \u2014 including charitable donations intended for the Congregation for the Evangelisation of Peoples \u2014 in a luxury Sloane Avenue development through a chain of offshore vehicles, exposed not bad actors so much as a legal architecture that is inherently opaque. Property held through thousands of juridic persons in dozens of legal systems, accumulating for centuries before modern accounting existed, resists not just disclosure but aggregation. The question of what the Church \"owns\" cannot be answered \u2014 not because the answer is hidden but because the question assumes a unit of ownership that does not exist.<\/code><\/pre>\n\n\n\nThe holdings include not just churches and convents but agricultural land \u2014 extensive in Latin America and sub-Saharan Africa \u2014 forest reserves, urban real estate, hospitals, universities, and colonial-era grazing land accumulated through donation, evangelisation, and land grants from governments that have long since ceased to exist. The diversity of asset types is itself part of the explanation for why no register is maintained: what would such a register even measure?<\/p>\n\n\n\n
What is clear, structurally, is this: the Church’s land was accumulated over centuries, which means it carries no acquisition cost in the present. It generates returns \u2014 rent, lease income, development gain \u2014 with no corresponding liability to shareholders, and, in many jurisdictions, with tax treatment that reflects the institution’s religious rather than commercial character. A landowner with no acquisition cost, limited tax exposure, and no obligation to report to anyone is not a landlord in the ordinary sense. The Church’s land is a different category of asset almost entirely, and the frameworks we use to track concentrated wealth \u2014 wealth rankings, shareholder registers, corporate filings \u2014 were not designed to see it.<\/p>\n\n\n\n
What is structurally significant here is not the acreage. It is the documented fact that the acreage cannot be established \u2014 that this unverifiability is not a limitation of the data but a property of the ownership itself. The largest landowners are invisible not because they hide but because they occupy a category that the instruments of financial legibility were never built to reach.<\/p>\n\n\n\n
The Crown That Is Not the King<\/h2>\n\n\n\n
The British Crown Estate holds approximately 116,000 hectares of agricultural land and forest in the UK, roughly half of the foreshore, and near-total rights over the seabed out to twelve nautical miles. Its offshore holdings extend further into the continental shelf. Its total estate was valued at approximately \u00a315.5 billion in its most recent annual report.<\/p>\n\n\n\n
None of this belongs to the King.<\/p>\n\n\n\n
The distinction is not a technicality. The Crown Estate belongs to “the Crown” \u2014 a constitutional abstraction that precedes and survives any individual monarch. The sovereign cannot sell it, does not pocket its profits (which flow to the Treasury, with a percentage returned to the royal household as the Sovereign Grant), and is not legally liable for its management. The entity is governed by Crown Estate Commissioners, appointed by the Crown on the advice of ministers, accountable to Parliament but not controlled by it, and operating outside the normal framework of both public sector accountability and private sector market discipline. It is too permanent and constitutionally embedded to be called private in any ordinary sense. Yet it is also too constitutionally insulated to be called public in any ordinary sense. It occupies the space between categories, which is exactly the space where the largest landowners tend to live.<\/p>\n\n\n\n
Seabed and the Clean Energy Dividend<\/strong>\nThe Crown Estate's offshore holdings acquired a new economic significance in the 2020s as offshore wind became central to UK energy policy. Because the Crown Estate owns the seabed, it owns the platform on which offshore wind turbines are built \u2014 and collects lease income from wind farm operators accordingly. The Crown Estate Act 2025 (which received Royal Assent on March 11, 2025, following introduction to the Westminster Parliament as the Crown Estate Bill in July 2024) granted the Crown Estate new borrowing powers specifically to facilitate offshore wind investment \u2014 a significant institutional expansion announced in the 2023 autumn statement. The Energy Act 2023 is a separate instrument and is not the relevant legislation for Crown Estate borrowing powers; Crown Estate Scotland, managed separately by Scottish Ministers following the Crown Estate (Scotland) Act 2019, operates under a further distinct framework. The question the Act quietly sidesteps is the same question the Church case raised: if the seabed beneath Britain's offshore wind installations belongs to a constitutional abstraction whose profits flow first to the Treasury, who is the real beneficiary of the clean energy transition? And by what mechanism, available to whom, could that question be answered?<\/code><\/pre>\n\n\n\nThe Windsor Estate \u2014 approximately 6,300 hectares \u2014 is a separate matter, representing something closer to personal royal property. It is also roughly one-twentieth the size of the Crown Estate’s agricultural holdings alone, which gives some indication of the scale difference between the monarchy’s personal wealth and the constitutional portfolio conducted in its name.<\/p>\n\n\n\n
What the Crown Estate shares with the Church’s dispersed holdings is the structural outcome: an enormous concentration of land that is not legible as private wealth, does not appear on wealth rankings, and cannot be challenged through the mechanisms available to voters or markets. Two different architectures \u2014 dispersal across thousands of juridic persons; consolidation within a constitutional abstraction \u2014 producing the same practical result.<\/p>\n\n\n\n
The Interior Empires<\/h2>\n\n\n\n
SLC Agr\u00edcola, controlled by the Logemann family and traded on the S\u00e3o Paulo stock exchange under the ticker SLCE3, operates approximately 661,000 to 830,000 hectares of planted area across multiple Brazilian states \u2014 a figure drawn from the company’s own investor materials and financial press reporting, though it represents operated and planted area rather than a simple measure of owned land. SLC combines owned holdings with leased and partnership-farmed arrangements; the owned portion is a subset the company does not always disaggregate publicly. The total operated area is roughly the size of the state of Delaware multiplied by five. The Logemann family built this through the Brazilian agricultural frontier \u2014 a process of expansion into the Cerrado and savanna that began with government subsidy, continued through land clearing, and was formalised into the kind of holdings that appear in investor presentations.<\/p>\n\n\n\n
The Bom Futuro Group, founded by Era\u00ed Maggi Scheffer and his brothers, operates more than 650,000 hectares of cultivated area across Mato Grosso and neighbouring states. Blairo Maggi \u2014 the billionaire former governor and senator known as the “Soy King” before he entered formal politics \u2014 is Era\u00ed’s cousin and heads the separately constituted Amaggi Group; the two operations represent different branches of the same extended family and different chapters of the same regional story. A third entity, the Bom Jesus Group, extends the same dynastic pattern across adjacent states \u2014 a further iteration of the same regional logic operating at comparable scale.<\/p>\n\n\n\n
The Cerrado's Invisible Crisis<\/strong>\nThe Brazilian Cerrado \u2014 the tropical savanna that covers roughly 25% of Brazil's land area \u2014 is the world's most biodiverse tropical savanna and the source system for eight of Brazil's major river basins. It has received a fraction of the conservation attention directed at the Amazon, despite having lost roughly half its native vegetation since the 1970s. The agribusiness expansion into the Cerrado was not accidental: government credit programmes under the PRODECER scheme (Brazil-Japan Agricultural Development Cooperation Programme) from 1979 onward subsidised the transformation of Cerrado land into soy and corn production at industrial scale, creating the economic basis for the dynastic holdings that now dominate it. The land being accumulated into these empires was not empty. It was ecologically occupied by one of the planet's most significant biomes, and culturally occupied by quilombola communities, indigenous groups, and smallholder farmers whose displacement is part of the same history that produced the 600,000-hectare portfolio.<\/code><\/pre>\n\n\n\nThese dynasties are legible in a way the Church and the Crown Estate are not. The Logemann family’s holdings are reported in investor materials; SLC Agr\u00edcola’s financials are public. That legibility has not produced any particular political challenge to the concentration \u2014 but it makes the contrast with what follows more precise. The Brazilian families are visible landowners by any reasonable definition, holding land in their names, publicly disclosed, traded on an exchange. If you wanted to know who owns the Cerrado, you could find out. The same cannot be said of the entities that have become the Cerrado’s new neighbours \u2014 not family names but fund vehicles, registered in jurisdictions that give no indication of who ultimately holds the asset.<\/p>\n\n\n\n
The Fund That Owns Your Pension’s Farm<\/h2>\n\n\n\n
Nuveen Natural Capital is the investment arm of TIAA \u2014 the pension manager for US university staff, teachers, and academic workers \u2014 and the largest manager of farmland globally. Its agricultural holdings amount to approximately 3 million acres under management across roughly $13.1 billion in farmland assets as of 2024. Macquarie Asset Management, the Australian financial services group, manages approximately 4.6 million hectares of farmland as of December 2024, with approximately A$4.7 billion in agricultural assets under management.<\/p>\n\n\n\n
Bill Gates, the most visible individual farmland owner in the United States at approximately 242,000 to 275,000 acres across roughly twenty states via his Cascade Investment vehicle, does not appear on these comparisons. He is dwarfed by Nuveen’s fund vehicles, and he is legible in a way that Nuveen’s holdings are not. Gates’s acreage appears in the Land Report 100 and has been confirmed by multiple investigative outlets. Nuveen’s 3 million acres is not a public fact in the same sense \u2014 it is an inference from regulatory filings, fund prospectuses, and asset management disclosures, none of which present it as a single consolidated figure.<\/p>\n\n\n\n
The actual entities in the institutional farmland space \u2014 Nuveen, Macquarie, Farmland Partners, Gladstone Land Corporation \u2014 are less recognisable names than the billionaires they displace from the top of any meaningful ranking. That gap between recognition and scale is not incidental to the point.<\/p>\n\n\n\n
The Post-2008 Asset Class<\/strong>\nFarmland became an institutional asset class through a specific sequence of events. The 2008 financial crisis destroyed confidence in financial instruments whose risks had been obscured by complexity; commodity prices spiked between 2007 and 2012, making food-producing land visibly valuable; and investment rationale began circulating within pension fund management circles \u2014 articulated publicly by TIAA-CREF in 2012 when it launched its Global Agriculture LLC vehicle \u2014 arguing that farmland provided inflation-hedging properties and low correlation with equity markets. The 2007\u20132008 food price crisis was a catalyst: when food prices spike, the asset underlying the food supply becomes a more attractive investment. By 2023, the IPES-Food Land Squeeze report documented 960 active funds specialised in food and agricultural assets, managing more than $150 billion, with global land prices having nearly doubled between 2008 and 2022 and tripled in Central and Eastern Europe. Institutional capital did not discover farmland by accident. It was directed there by crisis and by a body of investment theory arguing, correctly, that land is the one input into food production that cannot be manufactured.<\/code><\/pre>\n\n\n\nThe distinction that matters here is not scale but character. A dynastic fazendeiro \u2014 whatever the circumstances of the original land acquisition \u2014 has an interest in the long-term productivity of the specific land he holds. His cost basis is effectively zero (the land was cleared by his grandfather’s generation), but his horizon is indefinite; the land will pass to his heirs. An investment fund holds land as an asset with a target return, a defined investment horizon, and an obligation to its limited partners. Its relationship to the land is instrumental rather than territorial. The distinction between holding land as a place and holding it as a position in an asset class is not merely philosophical \u2014 it has consequences for how the land is managed, what tenants can expect, and what happens to local communities when the fund’s target return is met and the portfolio is rotated.<\/p>\n\n\n\n
Those consequences are documented. A 2023 Journal of Economic Geography paper on financialised agri-corporate investors in Australian agricultural regions between 2004 and 2019 revealed that these investors acquired almost 200,000 hectares of farmland through three distinct pathways \u2014 takeovers of listed agricultural companies, purchase of already-assetised farm portfolios, and aggregation of family farms \u2014 revealing what the authors describe as “flexible and opportunistic” dynamics in how financial capital amasses land assets and restructures regional agricultures. The downstream effects of this restructuring \u2014 pressure toward short-term extraction, reduced investment in soil health, constraint on local supply chains, indifference to existing operators \u2014 are documented separately in the IPES-Food Land Squeeze report, which traces how the incentive model of financial land ownership produces outcomes systematically different from those of conventional agricultural operations. These are not individual failures of management. They are the structural outputs of a specific incentive model applied to a fixed asset.<\/p>\n\n\n\n
The TIAA case makes the irony precise. TIAA is not an oligarch. It is a not-for-profit institution managing the retirement savings of American academics and teachers. Millions of people who have spent careers thinking about inequality and the structure of economic power are invested, through their pension accounts, in farmland portfolios that a 2020 Chain Reaction Research report documented as linked to fires, land conflicts, and legacy deforestation in Brazil. The report identified 2,970 hectares cleared between 2009 and 2018 on six TIAA-affiliated farms in Brazil, with 2,350 hectares burned in August 2019. The beneficial owners \u2014 the schoolteachers \u2014 have no awareness of the specific landholdings their retirement accounts represent, no control over the decisions made about those holdings, and no meaningful ability to contest them. The ownership is not hidden. It is structured so as to make the connection between the beneficiary and the land invisible by default.<\/p>\n\n\n\n
The pension fund disperses ownership through LP structures and SPVs \u2014 limited partnerships and special purpose vehicles whose names give no indication of their ultimate beneficial owner \u2014 in exactly the way the Church dispersed it across thousands of canonical entities and the Crown Estate consolidated it within a constitutional abstraction. Different legal architectures, different eras, the same structural outcome: an ownership form that evades the frameworks used to track concentrated economic power.<\/p>\n\n\n\n
The Logic That Makes It Permanent<\/h2>\n\n\n\n
David Ricardo identified the specific problem with land in 1817, in On the Principles of Political Economy and Taxation<\/em>, and the analysis remains the cleanest account of why this particular form of wealth concentration is qualitatively different from every other.<\/p>\n\n\n\nLand is fixed in supply. Unlike capital, it cannot be manufactured; unlike labour, it cannot be multiplied. This means that rent \u2014 the return to land as a factor of production \u2014 is not a reward for productive activity. It is a toll extracted from everyone who needs land and does not own it. When population increases or economic activity intensifies, the demand for land rises. The owners of land capture the entirety of that increase without having improved the asset. Ricardo called this “differential rent” \u2014 the surplus that flows to landowners simply because the land exists and demand for it grows. The mechanism is not exploitative in the classical sense; it requires no particular malice or even intent. It is a structural feature of an asset whose supply cannot respond to demand.<\/p>\n\n\n\n
The IPES-Food Land Squeeze report, published in May 2024, provides a recent quantification of where this logic has arrived: the top 1% of farms globally now control more than 70% of the world’s farmland. The International Land Coalition’s 2020 “Uneven Ground” report, produced in collaboration with Oxfam, supplies a different but complementary measure \u2014 the top 10% of the rural population captures 60% of agricultural land value, while the bottom 50% of rural populations capture only 3%. Not land area, but value \u2014 which is the more telling figure, since it accounts for the concentration of the most productive and commercially accessible land within the top tier.<\/p>\n\n\n\n
The housing consequence follows from the same mechanism, though the chain of causation is more politically contested. When land near employment centres is held in concentrated portfolios \u2014 whether dynastic, institutional, or constitutional \u2014 supply is constrained not because construction is difficult but because landowners benefit from scarcity. The UK Competition and Markets Authority’s final report on the housebuilding market, published in February 2024, documented land banking behaviour and identified concentrated ownership in land markets as a structural feature of the system, while concluding that land banking was a symptom rather than a primary cause, with the planning system as the upstream constraint. The mechanisms are real; the planning system sits upstream; the question of relative magnitude between the two remains open.<\/p>\n\n\n\n
The food system argument is more direct and better documented. The IPES-Food Land Squeeze report traces how financialised land ownership changes the conditions under which food is produced at a structural level. When land is managed by an investment fund with a defined return target, long-term soil health, local supply chains, and smallholder access are not variables in the model. The 960 active agricultural funds managing $150 billion as of 2023 are not farming operations; they are financial instruments that happen to hold a farming asset, and the distinction matters for what the land produces, for whom, and on what terms.<\/p>\n\n\n\n
The self-reinforcing feature is the one Ricardo identified: because land cannot be created, those who hold it during periods of rising demand capture appreciation without any improvement to the asset, and that capital can be deployed to acquire more land, at prices that smaller buyers cannot meet. Concentration does not merely persist; it compounds. The International Land Coalition documents increasing concentration in almost all world regions since 1980. The specific acceleration since the 1990s has two distinguishable causes: the liberalisation of agricultural land markets in the Global South under structural adjustment conditions attached to IMF and World Bank lending, which opened markets previously inaccessible to foreign capital; and the post-2008 financialisation of farmland as an asset class, which provided both the institutional capital and the theoretical legitimation for investment at unprecedented scale and speed. The ownership logic is not new. The pace at which it is operating is.<\/p>\n\n\n\n
The Visibility Problem<\/h2>\n\n\n\n
England and Wales’s Land Registry is among the more transparent land ownership systems in the world, which is a description that requires immediate context. A 2019 investigative book \u2014 Guy Shrubsole’s Who Owns England?<\/em> \u2014 required years of work reconstructing a picture of concentrated ownership from registry data. Shrubsole found that approximately 25,000 landowners hold roughly half of England; that the aristocracy and landed gentry retain approximately 30% of England despite everything the twentieth century did to redistribute it; and that roughly 17% of English land ownership remains entirely undeclared at the Land Registry. This is the transparent system.<\/p>\n\n\n\nIn most of the world, the situation is considerably worse. Land ownership data is fragmented across local registries, held in formats resistant to aggregate analysis, maintained with varying degrees of accuracy and completeness, and \u2014 crucially \u2014 not required to identify the ultimate beneficial owner of land held through corporate vehicles. A pension fund’s farmland holdings do not appear on any list of landowners. They appear in fund prospectuses, limited partnership filings, and land registry entries under vehicle names that give no indication of who ultimately holds the asset. Nuveen’s 3 million acres is not a fact that can be looked up. It is reconstructed from regulatory filings by researchers and journalists who are specifically looking for it, using a combination of legal entity searches, agricultural land register requests, and cross-referencing of corporate ownership chains that is methodologically laborious and incomplete by design.<\/p>\n\n\n\n
Each of the ownership forms traced here shares, as a structural feature, the capacity to prevent aggregate landholding from becoming a legible political fact. The Church holds land through thousands of canonical entities that predate modern accounting and resist consolidation. The Crown Estate holds it through a constitutional form that sits between public and private without being either. The Brazilian dynasties hold it in legible family names, but in a legal and political environment where legibility has not produced accountability. The pension funds hold it through LP and SPV structures that are opaque by architectural design. Three different architectures, three different centuries \u2014 the juridic persons of medieval canon law, the constitutional consolidation of the post-Restoration settlement, the LP structures of post-2008 asset management \u2014 producing the same practical outcome.<\/p>\n\n\n\n
This is not a conspiracy. The Catholic Church did not design canon law in 1983 in order to obscure its land from financial journalists. The Crown Estate’s constitutional ambiguity is a medieval inheritance, not a deliberate obfuscation. But the effect is identical: ownership at a scale that would be politically significant, if it could be seen, cannot be seen \u2014 not because anyone is hiding it, but because the category in which it exists falls outside the frameworks designed to make economic power visible.<\/p>\n\n\n\n
The political economy of this invisibility has a consistent structure across each case. The Church’s land is not redistributed because it is held by a religious institution, and the historical and spiritual legitimacy attached to that institution functions as an immunity. The Crown Estate is not challenged because it has always existed and because the alternative \u2014 “nationalising” the seabed \u2014 sounds more radical than it is, in a constitutional arrangement where the seabed has never been private. The Brazilian dynasties are not challenged because their holdings are produced within a system of agricultural capitalism that the country’s political economy depends on and, during commodity booms, celebrates. Nuveen’s farmland is not questioned because it is a pension fund, and questioning it appears to threaten teachers’ retirements \u2014 an immunity that may be the most ingeniously structured of all, since it positions the middle-class beneficiary as a hostage of the very concentration they might otherwise object to.<\/p>\n\n\n\n
These immunities are not accidental. They are features of the ownership structures, not defects in the public’s willingness to see them clearly. Each of these ownership forms has built its own case for why its land is different \u2014 exempt from the normal calculus of concentrated wealth, located in a category where the usual questions do not apply. Before any political response to land concentration is possible, the prior condition is legibility: the ability to see the concentration clearly, to name who holds what, to follow the ownership chain from the registered entity back to the ultimate beneficiary. That prerequisite is absent in almost every jurisdiction in the world, almost entirely in the categories where concentration is highest.<\/p>\n\n\n\n
Zamkni\u0119cie<\/h2>\n\n\n\n
The reader who, at the start of this article, would have named a Saudi prince or a tech billionaire has not merely received a corrected list. The correction is less interesting than its cause. The largest landowners are invisible not because they are obscure but because they occupy categories \u2014 religious institution, constitutional abstraction, family dynasty, fiduciary intermediary \u2014 that were designed, or have evolved, to sit outside the frameworks we use to see and contest concentrated economic power. Land is the oldest form of wealth, the least mobile, the most tangible. It is also, in an era that has produced more financial transparency than any in history, among the least visible categories of asset when measured by the distance between its actual concentration and its political legibility.<\/p>\n\n\n\n
Ricardo understood in 1817 why this matters more for land than for any other asset: a fixed supply means that concentration does not merely persist but compounds, and that the compounding requires no active effort from the owner, only the structural fact of ownership itself. What has changed in two centuries is not the logic but the architecture. The new enclosures are conducted through LP structures and offshore vehicles rather than Acts of Parliament, and the enclosers include teachers’ pension funds alongside colonial churches. The mechanism is unchanged. The invisibility is more sophisticated.<\/p>\n\n\n\n