Note: This is a research note supplementing the book Unscarcity, now available for purchase. These notes expand on concepts from the main text. Start here or get the book.

The Land Tax That Funds Abundance: Henry George’s 1879 Solution to a 2030 Problem

Or: The one tax base that AI can’t automate, capital can’t offshore, and corporations can’t avoid—and why it’s been suppressed for 145 years.

The Revenue Cliff Nobody’s Talking About

Everyone’s debating what AI will do to jobs. Almost nobody is asking what AI will do to government revenue.

Here’s the arithmetic. The U.S. federal government collected $4.9 trillion in revenue in 2024. Roughly half came from individual income taxes—taxes paid by people with jobs. Another 9% came from corporate income taxes. Payroll taxes (Social Security and Medicare) contributed 36%—again, taxes on people working.

Now apply the labor cliff projections. If structural unemployment reaches 30% by 2035, income tax revenue doesn’t decline by 30%. It declines more, because the highest-earning jobs (finance, law, software engineering) are among the most automatable, and high earners contribute a disproportionate share of income tax revenue. Payroll tax revenue collapses in tandem. Corporate profits may rise—but capital is mobile, and corporations have spent four decades perfecting the art of jurisdictional arbitrage.

The result: governments lose their revenue base at exactly the moment they need to spend the most—on displaced workers, social stability, and the infrastructure of transition.

This is the revenue problem that the Bootstrap Paradox identified as Mechanism 2. And it has an answer—one that a failed San Francisco mayoral candidate proposed in 1879, that a Nobel laureate proved mathematically optimal in 1977, and that the coming automation revolution makes more relevant than at any point in human history.

Henry George and the Book That Almost Changed Everything

In 1879, a self-educated journalist named Henry George published Progress and Poverty. It sold millions of copies—by some accounts, outselling every English-language book except the Bible during the 1890s. George was described as the “third most famous American” of his era, after Mark Twain and Thomas Edison. John Dewey estimated in 1933 that it “had a wider distribution than almost all other books on political economy put together.”

George’s argument was deceptively simple. He observed that as economies advance, land values rise—capturing the gains of progress for landowners while workers see no benefit. A railroad comes through town. The railroad workers don’t get rich. The people who owned the land along the tracks do. A city grows prosperous. The workers who built the prosperity don’t capture it. The landlords do.

His remedy: a single tax on land values to replace all other taxes.

The logic:

Land is not produced by human effort. Nobody built the land. Its value derives from location, community investment, and natural features—none of which the landowner created.
Taxing land doesn’t reduce its supply. Tax buildings, and people build fewer buildings. Tax income, and people work less (or hide income). Tax land, and… the land is still there. You can’t build less land. You can’t offshore it. You can’t hide it in a shell corporation in the Caymans. This makes land tax the only tax with zero deadweight loss—it doesn’t distort economic behavior.
Land value is community-created. A vacant lot in Manhattan is worth $30 million not because the owner did anything to it but because eight million other people built a city around it. The community created the value. The community should capture the value.

George nearly won the 1886 New York City mayoral election, finishing second with 68,110 votes—ahead of Republican candidate Theodore Roosevelt (60,435 votes) and behind Democrat Abram Hewitt (90,552 votes). Allegations of electoral fraud persisted. Roosevelt himself estimated 15,000 Republican votes swung to Hewitt specifically to block George. The Democratic and Republican establishments collaborated against his candidacy.

George Bernard Shaw, Leo Tolstoy, H.G. Wells, Friedrich Hayek, Albert Einstein, and Sun Yat-sen all cited George as a formative influence. In the UK, a vast majority of both socialist and classical liberal activists traced their ideological development to his work.

And then… the idea was quietly buried. For reasons we’ll explore.

The Henry George Theorem: When a Nobel Laureate Proved Him Right

In 1977, Joseph Stiglitz—who would win the Nobel Prize in Economics in 2001—published a formal proof of what he called the Henry George Theorem.

The theorem states: In a city of optimal population size, aggregate land rents equal expenditure on public goods.

Read that again. In an optimally sized city, the total rent paid for land exactly equals the total cost of public services. A single tax on land rents would be both efficient (zero distortion) and sufficient (fully funds public goods).

Arnott and Stiglitz (1979) formalized this further in the Quarterly Journal of Economics, establishing a “golden rule of local public finance.” Subsequent generalizations extended the theorem to conditions including congestion externalities, heterogeneous agents, multiple regions, monopolistic competition, and dynamic settings—the theorem holds even in present-value terms.

The theorem has been robust to every extension tested. It holds with congestion, with different populations, with uncertain outcomes. It’s one of the most durable results in public finance theory.

And yet: almost no jurisdiction in the world has implemented a pure land value tax.

Where It’s Been Tried: The Global Evidence

The evidence isn’t purely theoretical. Multiple jurisdictions have implemented partial land value taxation, with instructive results.

Singapore: The 90% model

Singapore is the closest thing to a functioning Georgist economy at national scale.

The government owns over 90% of all land—up from 44% in 1960 to 76% by 1976, continuing to increase. Housing and Development Board (HDB) flats are sold on 99-year leaseholds; the land remains government property and reverts to the state when the lease expires.

The results:

Homeownership rate: ~89-90%—among the highest in the world
Even the bottom 10% income bracket has 84% homeownership
Government Land Sales averaged $5.5 billion annually (FY2017-FY2021)
GDP per capita exceeds $90,000—one of the world’s highest

Singapore demonstrates that public land ownership and near-universal homeownership are not contradictory. The government captures the land value; citizens own the improvements (their flats). This is Henry George’s principle in practice: the community captures community-created value, while individuals retain the fruits of their own investment.

Important structural detail: GLS proceeds accrue to Past Reserves and are invested for long-term returns—they’re not available for current spending. This creates an endowment model rather than a revenue-dependent one.

Hong Kong: The cautionary evolution

Hong Kong operated under a similar model. All land was government-owned and leased to private users. From 1970 to 1991, land revenues accounted for approximately 20% of the total government budget.

But the model has weakened. After 1997, land premium revenue became volatile and less dominant. In 2023-24, projected land premium of HK$85 billion came in at just HK$19.4 billion due to property market downturn. In 2024-25, Financial Secretary Paul Chan publicly stated Hong Kong would reduce dependence on land sales.

The lesson: Land revenue works brilliantly during rising markets. During downturns, it can crater. A pure LVT (annual rent capture) is more stable than a land sales model (one-time lease auctions), because rents track value continuously rather than depending on discrete transaction events.

Estonia: The incomplete experiment

Estonia implemented a land tax on May 10, 1993, as part of post-Soviet reform. Tax rates are set by local councils within 0.1% to 2.5% of assessed land value. Land tax accounts for approximately 7% of local revenues.

The results are mixed. Land tax revenues increased 171% between 1994 and 1999 due to revaluations and improved collection. But the assessment was set in 2001 and never updated. By 2003-2008, the gap between assessed and market values had grown enormous—and Estonia experienced a massive housing bubble despite the land tax.

The lesson: A land tax without regular reassessment is worse than useless. It creates a false sense of policy adequacy while the actual market diverges from the tax base. Estonia updated its assessments effective January 2026—25 years late.

Harrisburg, Pennsylvania: The American success story

Harrisburg began shifting to a split-rate property tax in 1975, reducing the rate on buildings while increasing the rate on land. By 1982, the tax rate on land was four times the rate on buildings.

The results by the late 1980s:

Vacant properties: reduced from over 5,000 to just a few hundred
Land and building values: increased from ~$400 million to $1.7 billion (inflation-adjusted)
Over 90% of property owners paid less than under the previous system
Municipal tax rates dropped throughout the 1990s while revenues stayed stable

Harrisburg demonstrates the land tax’s most powerful behavioral effect: it makes holding vacant land expensive, which forces development or sale. When speculators can’t profit from holding empty lots, they build on them or sell to someone who will.

Pittsburgh: The cautionary American tale

Pittsburgh used a split-rate (graded) tax from 1913 to 2001—nearly 90 years, with “considerable evidence of its success, especially in the first decades.”

It was abandoned in 2001 due to a reassessment crisis. The first real market valuation in decades caused land taxes to jump 81% and building taxes by 43%. Affluent residents filed thousands of appeals.

The lesson: The problem wasn’t the split-rate concept. The problem was decades of deferred reassessments creating a false baseline. When the tax base was finally updated to reflect reality, the adjustment was politically catastrophic. Pittsburgh is “a compelling example of the limited role that evidence often plays in policy decisions.”

Denmark and Taiwan: Other models

Denmark levies a land value tax (grundskyld) on all privately owned land, accounting for approximately 13% of municipal revenue. Values are assessed every two years—the regular reassessment that Estonia lacked.

Taiwan has a constitutionally mandated Georgist influence, thanks to Sun Yat-sen’s direct encounter with Henry George’s ideas. Taiwan’s Land Value Increment Tax charges 20-40% on the increase in government-assessed land value during ownership—a capital gains tax specifically on land appreciation. It was designed to curb speculation, though its share of national revenue has declined from 13% to under 7% since 2003.

Why LVT Becomes Critical in a Post-Labor Economy

Here’s where the 145-year-old idea meets the 2030 reality.

The automation tax base problem

Acemoglu, Manera, and Restrepo (2020) found that the U.S. tax code already favors automation: the effective tax rate on capital invested in equipment and software has declined to about 5%, while effective labor taxes stand above 28.5%. This disparity encourages firms to automate beyond the socially optimal level. Moving to optimal taxation would raise employment by 4.02%.

The asymmetry will only worsen. As AI and robots replace more workers, the tax base shifts from heavily-taxed labor to lightly-taxed capital. Governments face a choice: tax capital more (and risk capital flight), tax consumption more (and crush the consumer economy), or find a tax base that can’t flee.

A January 2026 Brookings working paper—“The Future of Tax Policy: A Public Finance Framework for the Age of AI”—explicitly identifies the solution: in an AI-driven economy, taxable sources of rent include “unimproved land, spectrum rights, unique datasets that cannot be replicated, and monopoly profits from AI market concentration.” The paper states:

“Identifying and taxing economic rents becomes more valuable as labor’s share declines, since rents—returns above what is necessary to induce an activity—can be taxed without distorting investment decisions.”

Land is the paradigmatic economic rent. No one created it. Taxing it doesn’t discourage any productive activity. And unlike spectrum rights or AI monopoly profits (which require complex regulatory apparatus), land is physically identifiable, locally governed, and impossible to move.

The Macroeconomic Henry George Theorem

Mattauch, Siegmeier, Edenhofer, and Creutzig (2018) published a critical extension: “Financing Public Capital through Land Rent Taxation: A Macroeconomic Henry George Theorem” in Public Finance Analysis.

Their proof: the socially optimal level of public capital can be reached by a land rent tax, provided land is a more important production factor than public capital. This avoids the trade-off between efficiency-enhancing public investment and market efficiency created by distortionary taxes.

The implication for the Unscarcity framework: as automation makes capital abundant and labor less central, land rents become a larger share of national income. The Henry George Theorem becomes more relevant at the macroeconomic level—not just the local public finance level.

Digital Georgism

Eric Posner and Glen Weyl’s Radical Markets (Princeton University Press, 2018) drew directly on George for their Common Ownership Self-Assessed Tax (COST)—a modern Harberger tax where property holders self-assess their assets and must sell at the self-assessed price if a buyer offers it.

Their broader insight extends George to the digital economy: “The most expensive real estate on earth is no longer 5th Avenue in Manhattan or Nanjing Lu in Shanghai. It’s pixels on the Google landing page.” Digital platforms create location-dependent rents analogous to physical land rents—dominant platforms extract monopoly rents from network effects. Digital Georgism proposes treating these digital monopoly rents the same way George treated physical land rents: as community-created value that should be captured for public benefit.

The Free Zone Application: LVT as Self-Funding Mechanism

In the Unscarcity framework, the Land Value Tax isn’t a policy proposal for existing governments (though it could be). It’s the self-funding mechanism for Free Zones.

Here’s how it works:

Phase 1: Foundation construction (Year 0-5)

During initial construction, land in the Free Zone has minimal value. The zone is a construction site in the Australian outback or central Kansas. Nobody wants to live there yet. Land tax revenue is negligible.

Funding source: Transition Trusts. The founding consortium’s $4-8 billion funds infrastructure construction. The LVT doesn’t fund the first shovel—this is the bootstrapping sequencing note that the original article acknowledged. You need external capital to build the thing that creates the land value that funds the ongoing operations.

Phase 2: Proof-of-concept (Year 5-10)

The first 5,000-50,000 residents arrive. Housing, food, healthcare, and energy work. The community is functional. Word spreads. Demand for residence increases.

As demand increases, the value of being in the zone—the locational value—rises. This is land value. A plot inside the functioning Free Zone is worth more than an identical plot 10 miles outside it, because the inside plot comes with access to Foundation services.

The LVT activates. The zone captures 100% of this locational value. No private land ownership—residents have secure tenure (you can’t be evicted arbitrarily) but the land rent flows to the Commons. This funds an increasing share of operational costs.

Phase 3: Self-sustaining (Year 10-20)

As the zone becomes genuinely desirable—perhaps the most desirable place to live within a 500-mile radius—land values rise substantially. The LVT generates enough revenue to cover all Foundation services without external funding. The zone is fiscally independent.

At this point, Transition Trust funding can redirect to building the next zone instead of subsidizing the first. The first zone’s LVT makes it self-sustaining. The Trust capital that built it is now freed for replication.

Why 100% capture (not partial)

Most real-world LVT implementations capture a fraction of land value. Singapore’s leasehold model, Hong Kong’s land premium, Estonia’s 0.1-2.5%—all leave substantial land rents in private hands.

The Free Zone captures 100% for a specific reason: in a post-scarcity community, private land ownership creates the feudal dynamics the framework exists to prevent.

If residents can buy and sell land in the Free Zone, early settlers capture the appreciation from community investment—the exact problem George identified in 1879. The first residents to own land become landlords. Late arrivals pay rent to early arrivals. Class stratification recreates itself.

Full capture prevents this. Every resident has equal access to location. Nobody profits from owning the ground beneath the community’s feet. The community’s investment in making the location desirable returns to the community, not to early speculators.

This is only viable because the Free Zone starts on empty ground. You’re not confiscating anyone’s existing property—you’re establishing a new governance structure on land that had no significant private claims. This sidesteps the transition cost problem that bedevils LVT advocacy in existing cities.

Why It Failed Politically (And Why the Free Zone Bypasses the Failure)

If the land value tax is so theoretically sound, why hasn’t it been adopted widely?

The failure of the land value tax comes down to five structural factors:

1. Concentrated losers, diffuse winners. Landowners who would pay more are politically organized and well-funded. Everyone else who would benefit is diffuse and unaware. This is the classic public choice problem—concentrated interests defeat diffuse benefits.

2. Expanding homeownership. When George wrote in 1879, a small number of large landholders owned most land. Today, homeownership rates in developed countries range from 60-90%. Tens of millions of voters own property and perceive LVT as a threat to their home values. “Politicians are reluctant to alienate” this massive voting bloc.

3. Assessment difficulty. Separating land value from improvement value is technically challenging. Current assessment practices “have no real incentive to get the breakdown of value between land and improvements right.” However, this objection may be overstated: “Experience in Australia, Chile, Jamaica, New Zealand, Uruguay, and many other countries refutes this claim.” Modern GIS, multivariate analysis, and satellite imagery make assessment increasingly feasible.

4. Transition costs. Current property prices capitalize the expectation of low land taxes. Introducing LVT would reduce land values, creating windfall losses for current owners. Whether to compensate them is an unresolved debate even among Georgists.

5. Status quo bias. “Changing tax systems is difficult because people are familiar with existing systems and nervous about change.”

Every one of these problems disappears in a Free Zone.

The Free Zone is built on empty ground—no existing landowners to compensate. There’s no homeowner voting bloc because there are no prior homeowners. Assessment isn’t an issue because the zone starts from scratch with purpose-built valuation systems. There’s no transition cost because there’s no transition—the LVT is the founding fiscal charter. And there’s no status quo to defend because the community is new.

This is why the Free Zone model matters for LVT specifically. The political economy that has blocked land value taxation for 145 years doesn’t apply when you build the community from the ground up. You’re not fighting existing interests. You’re creating new institutions on blank canvas.

The Bootstrapping Sequencing Problem

The Bootstrap Paradox article acknowledged a sequencing challenge: the LVT doesn’t fund the first shovel. You need Transition Trust capital to build the infrastructure that creates the land value that the LVT captures.

But the sequencing is more nuanced than “external funding first, LVT second.” It’s a graduated handoff:

Year	Transition Trust funding	LVT revenue	External dependency
0-3	100%	~0%	Total
3-5	80%	20%	High
5-10	40%	60%	Moderate
10-15	10%	90%	Low
15+	0%	100%+ (surplus)	Self-sustaining

The “100%+” in the final row is intentional. A thriving Free Zone with 50,000+ residents generates land value that exceeds Foundation operating costs. The surplus can fund replication—seeding the next zone’s construction, creating the exponential growth the framework requires.

This graduated handoff is what distinguishes the model from donor-dependent communities. Masdar City, Auroville, and other intentional settlements remain dependent on external funding indefinitely because they never developed an internal revenue mechanism tied to their own success. The LVT creates exactly this feedback loop: as the zone succeeds, it generates the revenue to sustain itself and replicate.

Objections and Honest Responses

“You can’t separate land value from improvement value”

This is the most common technical objection. It has some validity in dense urban settings where land and buildings are deeply intertwined. But it has less validity in a Free Zone context because:

The zone starts on undeveloped land with clear baseline values
Improvements are largely standardized (modular housing, common infrastructure)
Modern assessment technology (GIS, satellite imagery, algorithmic valuation) handles the separation increasingly well
In a Foundation economy where housing is provided, the “improvement” component is communal rather than individual—simplifying the assessment

“100% capture kills incentive to improve”

If the zone takes 100% of land rent, what incentive does a resident have to maintain or improve their space?

The answer: personal improvements to buildings (not land) are not captured. If you renovate your kitchen, build a garden, or add a workshop to your dwelling, you keep the full value of that improvement. The LVT captures the locational value—the value of being in this community rather than somewhere else. It doesn’t capture the value you personally add to your dwelling.

This is the standard Georgist distinction: tax the land, not the building. It’s already how Harrisburg’s split-rate works—and Harrisburg saw more development after shifting to land taxation, not less.

“What about commercial land within the zone?”

Internal businesses—workshops, restaurants, co-working spaces—would pay land rents proportional to the locational value of their sites within the zone. Prime locations (near the community center, transit hubs) pay higher rents than peripheral locations. This mirrors how commercial real estate works in any city—but the rent flows to the Commons rather than to private landlords.

“This only works at Free Zone scale—it doesn’t help the billions outside”

Correct. The LVT is the self-funding mechanism for Free Zones, not a global solution. For the billions who remain in traditional economies during the transition, the Bootstrap Paradox identified Deflationary QE as the bridge mechanism. The LVT proves the model; scaling to the broader world requires existing governments to adopt LVT principles—something that becomes politically easier once Free Zones demonstrate it working.

What This Means for the Framework

The Land Value Tax is the second of four mechanisms in the Bootstrap Paradox solution. Its role is specific and critical:

Without LVT: Free Zones depend on Transition Trust funding forever. This creates the donor-dependency pattern that has killed every intentional community in history. When the funding stops, the community dies.

With LVT: Free Zones become self-sustaining once their locational value exceeds their operating costs. Trust capital that built the first zone is freed to build the second. The model replicates without requiring infinite external funding.

The theoretical case is 145 years old and mathematically proven. The practical case is demonstrated in Singapore, Harrisburg, and Denmark. The political obstacles that have blocked broader adoption don’t apply to greenfield Free Zones.

Henry George couldn’t implement his vision in 1886 New York because the city was already full of landowners. The Free Zone starts empty. The land tax isn’t imposed on existing interests—it’s the founding charter of a new community, agreed to by everyone who joins.

George’s 1879 solution fits the 2030 problem better than it fit the 1879 problem. When robots do the work and capital crosses borders at the speed of light, land is the one factor that stays put, stays scarce, and stays taxable. It’s the fiscal bedrock of a post-labor economy—available to any community willing to build on it from scratch.

References

Henry George, Progress and Poverty (1879) — The original land tax thesis
1886 New York City Mayoral Election - Wikipedia — George vs. Roosevelt vs. Hewitt
Stiglitz, J.E. (1977). “The Theory of Local Public Goods” — The Henry George Theorem
Arnott, R.J. and Stiglitz, J.E. (1979). “Aggregate Land Rents, Expenditure on Public Goods, and Optimal City Size” — Quarterly Journal of Economics
Mattauch et al. (2018). “Financing Public Capital through Land Rent Taxation” — Macroeconomic Henry George Theorem
Dynamic Henry George Theorem (2025) — Extension to dynamic settings
Public Housing in Singapore - Wikipedia — 90%+ government land ownership
Singapore Ministry of Law: GLS Revenue — Land sales data
Hong Kong: Major Sources of Government Revenue - LegCo — Land premium history
Hong Kong to reduce dependence on land sales - SCMP — 2024 policy shift
Land Reform and Taxation in Estonia - Lincoln Institute — Post-Soviet implementation
Boom and bust in Estonian real estate - ScienceDirect — LVT and the 2003-08 bubble
How Harrisburg was Transformed - Labour Land Campaign — Split-rate success story
The Pittsburgh Experience - Land Value Tax Guide — Reassessment crisis
Non-Glamorous Gains: Pennsylvania - Strong Towns — Broader state experience
Henry George and Sun Yat-sen - Substack — Taiwan’s Georgist constitutional influence
Acemoglu et al. (2020). “Does the US Tax Code Favor Automation?” — 5% capital vs. 28.5% labor tax rates
Brookings (2026). “The Future of Tax Policy: A Public Finance Framework for the Age of AI” — Rent taxation in AI economy
Posner and Weyl (2018). Radical Markets — COST proposal and digital Georgism
The Failure of the Land Value Tax - Works in Progress — Political economy analysis
If the Land Tax Is Such a Good Idea - Strong Towns — Adoption barriers
Land Value Tax - D.C. Policy Center — Assessment challenges
Acemoglu and Restrepo (2019). “Automation and New Tasks” — Displacement/reinstatement framework
Unscarcity, Chapter 8: The Transition — Full narrative context

Land can’t be automated. Land can’t be offshored. Land can’t be hidden in a shell corporation or routed through a tax haven. In a world where everything else becomes abundant, location remains scarce. Henry George saw this in 1879. Stiglitz proved it in 1977. The automation revolution is about to make it the most important insight in public finance. The only question is whether we apply it—to Free Zones first, and then everywhere that chooses to follow.

The Land Tax That Funds Abundance: Henry George's 1879 Solution to a 2030 Problem