Hudson: 'How to Lie with Real Estate Statistics' and 'Where Did All the Land Go?' (2001)

Overview

In two companion 2001 essays — "Where Did All the Land Go? The Fed's New Balance Sheet Calculations: A Critique of Land Value Statistics" and "How to Lie with Real Estate Statistics" — Michael Hudson documents how the US Federal Reserve's land-valuation methodology produced a negative $4 billion value for all land owned by non-financial US corporations in 1993, an absurdity Hudson argues led the Fed to stop publishing comprehensive land-and-building breakdowns of real estate value after 1994.[1] The essays formed the basis for a colloquium at New York University's Real Estate Institute (October 25, 2001), co-organized by the Robert Schalkenbach Foundation and building on earlier unpublished research for the Foundation.[1] Hudson argues the Fed's "land-residual" method — subtracting the estimated replacement cost of buildings from overall market price to back into a land value — systematically understates land value wherever buildings are expensive to replace relative to what the market will actually pay, as in dense, heavily built-up urban sites such as Lower Manhattan.[1] Hudson estimates his own approach raises the Fed's 1994 land valuation by as much as $4.5 trillion.[1]

Findings

• The Fed's 1994 estimate put total US land value at $4.4 trillion against $9 trillion in building value; Hudson's own historically based estimate reverses the ratio, putting land at roughly two-thirds of total real estate value rather than one-third.[1]
• The land-residual method assigns negative values to heavily built-up sites — on Hudson's account, including the land under Manhattan skyscrapers such as the pre-9/11 World Trade Center — whenever the cost of rebuilding a high-rise exceeds the site's market price, an artifact of the accounting method rather than a fact about the site's economic value.[1]
• Hudson connects the statistical distortion to tax policy: because land-price gains are recorded as capital gains on buildings rather than land, and because buildings can be re-depreciated at each resale, "a substantial portion of the rise in site value is treated as depreciable building value, not as non-depreciable land-price gain," letting real estate investors shelter land-value gains from taxation.[1]
• Lars Doucet's Land Is a Big Deal (2022) independently cites the same negative-$4-billion episode (Ch. 14) as a key illustration of how the cost-based (Fed) approach undervalues land relative to hedonic-regression and vacant-land-sales methods, which put 2020 US land value in the $22–65 trillion range.[2]

Limits and Caveats

The 2001 essays are published on Hudson's own site/professional archive rather than in a peer-reviewed journal, and the $4.5 trillion revaluation figure is Hudson's own estimate rather than an independently replicated econometric result, and no independent replication of that figure was located this session; it should be read as Hudson's estimate, not a confirmed result. The core empirical claim — that the Fed's land-residual method produced a negative $4 billion land value for non-financial corporate land in 1993, and that the Fed subsequently narrowed its published real-estate statistics — is corroborated by Hudson's own detailed walk-through of Federal Reserve flow-of-funds data and is cited independently in Doucet's book.[1][2]

See Also

• Michael Hudson — author page
• Land Is a Big Deal (book) — independently cites the same $4 billion negative-land-value episode (Ch. 14)
• Larson, US Land Value — a rival, hedonic-regression-based estimate of US land value
• FIRE Sector — the finance-insurance-real-estate framework Hudson applies
• Economic Rent

Sources

1. Michael Hudson, "How to Lie with Real Estate Statistics," michael-hudson.com, 25 March 2001 (essay basis for an October 25, 2001 colloquium at NYU's Real Estate Institute, co-organized by the Robert Schalkenbach Foundation); companion essay "Where Did All the Land Go? The Fed's New Balance Sheet Calculations: A Critique of Land Value Statistics," michael-hudson.com, March 2001 — used for the Fed methodology critique, the negative $4 billion 1993 figure, the Fed's narrowing of the land/building series after 1994, the $4.5 trillion revaluation estimate, and the Manhattan/high-rise land-residual examples (A-claim; primary source, full text read). Article · Companion essay
2. Lars Doucet, Land Is a Big Deal (2022), Ch. 14 — used for independent citation of the same $4 billion negative-land-value episode and its place in the wider land-valuation-methods literature (B-claim). Wiki book page