AEI Land Price and Land Share Indicators (dataset)

Summary

The AEI Land Price and Land Share Indicators are a US dataset published by the American Enterprise Institute Housing Center that estimate, for the residential stock, what share of a home's value is land rather than structure, down to the census-tract, ZIP-code, county and metro level.[1] Lars Doucet uses the dataset in Does Georgism Work? Part 1 as the empirical backbone of his claim that most urban property value is land: he reads San Francisco County at roughly 70.9% land share (2020), congruent with his own vacant-lot "spot check" of ~76%.

The dataset is not a fresh valuation study but a repackaging of a peer-adjacent working paper already on the wiki. The AEI page states its indicators "are based on data for 2012 from The Price of Residential Land for Counties, ZIP Codes, and Census Tracts in the United States" by Davis, Larson, Oliner and Shui — the FHFA/AEI land-price paper the wiki already covers.[1][2] The AEI indicators extend and update that base with an "AEI-adjusted" methodology.

Method (residual land share)

The methodology document is explicit that land is estimated as a residual: "Knowing the market price allows us to estimate the land price, as the residual. The land share ratio may now be calculated."[2] Structure value is assumed to grow with inflation ("we assume that the structure value increases at the rate of inflation as the physical structure does not change much"), and depreciation/obsolescence is "presumed to be small and therefore not taken into consideration."[2] AEI motivates the focus on land by noting land is "the much more volatile component" of home value and that a rising land share unsupported by utility was, in prior work, "the 'canary in the coal mine'" for the 2000s housing bust.[2]

What It Supports / Limits

• Supports "most urban value is land." The dataset is the fine-geography evidence behind Doucet's headline that in expensive coastal metros the land share of home value is very high — the SF figure being the load-bearing example.
• Residual-method caveat (important). Because land = price − (inflation-indexed) structure, the estimate inherits any error in the structure assumption; AEI's own note that depreciation is ignored means structure value can be overstated and land share understated in older stock, or the reverse where the inflation assumption is wrong. This is the same residual-method limitation the mass-appraisal-methods page flags for assessors.
• Not independently re-verified here. The specific 70.9% San Francisco County figure comes from the AEI interactive tool as Doucet read it; it is a live-dataset value, not a static quotation, and is recorded here as Doucet's cited number rather than one this page re-derives.

Sources

1. AEI Housing Center, Land Price and Land Share Indicators. aei.org/housing/land-price-indicators — used for the existence, geography (tract/ZIP/county/metro) and provenance of the dataset, and its statement that the indicators are based on 2012 data from the Davis–Larson–Oliner–Shui residential-land paper (verified 2026-07-11).
2. AEI Housing Center, AEI-adjusted Land Price and Land Share Indicators Methodology (updated May 2021). PDF — used for the residual land-price method, the inflation-indexed structure assumption, the ignored-depreciation caveat, and the "canary in the coal mine" rationale. Quotes verified verbatim against the PDF (2026-07-11).
3. Morris A. Davis, William D. Larson, Stephen D. Oliner & Jessica Shui, The Price of Residential Land for Counties, ZIP Codes, and Census Tracts in the United States — the underlying working paper, summarized on the wiki as Davis, Larson, Oliner & Shui (FHFA).

See Also

• Doucet, Does Georgism Work? — Part 1, the 70.9% SF land-share citation
• Davis, Larson, Oliner & Shui (FHFA land prices) — the underlying paper
• What's Manhattan Worth? (Barr, Smith & Kulkarni) — a complementary land-magnitude data point in Part 1
• Saez & Zucman, Wealth Inequality in the US since 1913 — the Part 1 concentration companion
• Mass Appraisal Methods — the residual method in assessment practice