— economics · land & tax
solving the 145-year assessment problem through the credible threat of destruction
since henry george in 1879, economists have agreed: tax land values, not improvements. land value tax (l.v.t.) eliminates deadweight loss, captures unearned rent, and can fund public goods and u.b.i. without punishing productivity.
so why hasn't it been implemented at scale?
the assessment problem
bureaucrats can't reliably separate land value from improvement value. assessments are expensive, imprecise, politically contentious, and generate endless litigation. without a clean way to measure land value alone, l.v.t. can't be implemented correctly.the solution isn't a better assessor. it's a mechanism that makes landowners reveal their own land values honestly — by making dishonesty costly.
the mechanism works as follows. a property owner declares a land value and an improvement value separately. the government taxes the declared land value. anyone can challenge the declared land value by offering to buy the bare land at that price.
if challenged, the owner faces a choice: accept the buyout of the land, or have the improvements destroyed and lose the land anyway. this isn't punitive — it's the mechanism that makes the declaration credible.
if you understate your land value, someone will call your bluff and buy it cheaply. if you overstate it, you pay too much tax. the only rational declaration is the true market value. the threat of destruction is what enforces honesty — not a tax rate, not an assessor.
note what this does not require: no bureaucratic appraisal, no appeals process, no politically-negotiated valuation. the market discovers the land value directly, because the owner has skin in the game on both sides of the number.
this mechanism is sometimes framed as "harberger taxation" and sometimes as a vickrey auction. the distinction is largely superficial, and understanding why clarifies the core logic.
owner self-assesses land value continuously. pays a percentage tax on that declared value. anyone can force a sale at the declared price at any time. incentive to declare honestly: overstate and you pay too much tax; understate and you lose the land.
periodic auction where bidders submit land values and improvement values separately. highest total bidder wins. second-price rule makes truthful bidding dominant. destruction fallback enforces honest component separation.
these look different but produce the same equilibrium. here's why: under harberger, if someone calls your underbid and ousts you, you can immediately bid back in at a higher price. the forced sale doesn't strand you — it just resets the declared value to market. the ouster is credible only because the threat is real, not because it actually happens in equilibrium. rational owners declare honestly and the mechanism never fires.
the auction format is an implementation question. the truth-revelation engine is the destruction threat, not the specific price rule.
because the tax falls only on land value — not improvements — you capture the full return on anything you build. adding $150k of improvements raises your property's value by $150k and your tax bill by zero. same marginal incentives as fee simple ownership.
when a subway gets built nearby and land values rise $1m, the owner's declared land value must rise or they face a cheap buyout. that rise flows to public revenue. land appreciation — which is socially created, not earned by the owner — goes to the public.
the owner does the assessment under adversarial conditions. anyone who thinks the declared value is too low can act on that belief. no appeals, no valuation disputes, no politically-captured assessors. the market discovers prices through the threat mechanism.
the historic barrier to l.v.t. was separating land value from improvement value. this mechanism forces that separation: the owner has to declare them independently, and the destruction fallback makes dishonest component allocation costly. the market does what bureaucrats couldn't.
| feature | traditional l.v.t. | fee simple | destruction-threat mechanism |
|---|---|---|---|
| captures land appreciation | ✓ | ✗ | ✓ |
| preserves investment incentives | ✗ | ✓ | ✓ |
| no bureaucratic assessment | ✗ | ✓ | ✓ |
| incentive-compatible disclosure | n/a | n/a | ✓ |
| implementation difficulty | very high | low (status quo) | medium |
if a developer values the bare land more highly than you declared, they can buy it at your stated price. you receive the declared land value plus your improvement value — the same exposure you have under fee simple if the market shifts to higher-value use. the mechanism doesn't create new risk; it replicates fee simple risk while redirecting land appreciation to public revenue.
because l.v.t. only works if you can tax land without taxing improvements. taxing improvements punishes productive investment — exactly the deadweight loss l.v.t. is designed to eliminate. forcing separate declarations under adversarial conditions is the only mechanism that produces honest component separation without bureaucratic appraisal.
continuous declaration (harberger style) or periodic auction rounds (vickrey style) both work. in the continuous model, the 30-year treasury rate converts between the flow value (annual rent) and the stock value (declared price). the math is econ 101; the implementation is a policy choice.
the destruction threat is bilateral. if a challenger lowballs a forced acquisition, the incumbent can re-enter the market immediately and bid back in — so a predatory ouster just resets the price. if an incumbent understates land value, any party who believes the land is worth more can profitably acquire it. both sides have adversarial incentives toward honesty.
gaffney (2009) estimates u.s. land value at $23 trillion. at 5% annual return: $1.15 trillion per year — roughly 6% of g.d.p. enough to fund substantial u.b.i., replace distortionary taxes on income and capital, or invest in public infrastructure.
this isn't primarily about land taxation. it's about mechanism design — using incentive structures to solve problems that bureaucratic systems can't. the 145-year failure of l.v.t. implementation wasn't a political failure; it was a measurement failure. bureaucrats genuinely cannot separate land from improvement value reliably.
the destruction-threat mechanism solves this by replacing administrative judgment with adversarial incentives. the owner knows what their land is worth. the mechanism just makes it rational to say so.
pure georgist l.v.t. with no bureaucratic assessment. government captures market land value. investment incentives fully preserved. the tools were available in 1961. the bottleneck was always the mechanism, not the math.