Taxes, Mortgage Borrowing, and Residential Land Prices

Summary

"Taxes, Mortgage Borrowing, and Residential Land Prices" by Dennis R. Capozza (University of Michigan), Richard K. Green (then Wisconsin), and Patric H. Hendershott (then Ohio State) is a chapter in Henry J. Aaron & William G. Gale, eds., Economic Effects of Fundamental Tax Reform (Brookings Institution Press, 1996).[1] It circulated in working-paper form as "Taxes, Mortgage Borrowing and House (Residential Land) Prices" (Wisconsin CULER 96-06), and the same research program produced the closely related "Taxes and House Prices" (1997).[2][3]

The paper builds a user-cost model of owner-occupied housing — in equilibrium the cost of owning must equal the cost of renting the same dwelling — in which the US tax code's preferences for homeownership (non-taxation of imputed rent; deductibility of mortgage interest and property taxes) lower the after-tax cost of ownership and thus raise what buyers will pay.[2] Estimating this across US metropolitan areas (a panel of 63 metros, 1970–1990 in the companion analysis), the authors test whether those tax preferences are capitalized into house prices — and find that they are, fully.[2][3]

The title's emphasis on residential land carries the incidence point: structures are reproducible at construction cost (elastically supplied), so tax-driven changes in housing demand are absorbed by the fixed factor — land — rather than by the price of structures. The tax preference is capitalized into the site, and its removal would fall on whoever owns the site at the time.

Core Findings

• Full capitalization of income-tax preferences into house prices. In the companion estimation the authors report that their evidence "is consistent with full capitalization of marginal income tax rates into house prices."[3] Richard Green's own account of the method notes that the estimated tax coefficients were "not statistically different from one," the model's prediction, from which they "concluded that taxes do get capitalized into house prices" and ran policy simulations on that basis.[2]
• Removing the deductions would cut housing values ~13–17%. Applying the model to proposed reforms, the paper computes "substantial house price declines and wide geographic variation."[3] As an independent secondary source summarizes it: "one study of 63 metropolitan areas estimates that eliminating the mortgage interest and property tax deductions would reduce housing values by an average of 13 to 17 percent (Capozza, Green, and Hendershott 1996)."[4]
• Wide geographic variation. Because the capitalized value of a tax preference is larger where the land share of value is larger, the predicted price declines concentrate in high-cost, land-scarce metros — the same land-share logic that runs through the land-scarcity literature.[2][3]

Why it Supports the Claim — and its Limits

Supports the capitalization pillar. The result is a demonstration that a tax variable is fully capitalized into house (residential land) prices, so its burden — or, symmetrically, its benefit — accrues to the owner at the moment of the change, not to future occupants. That is the same asset-price incidence mechanism the wiki's landlords-cannot-pass-LVT claim relies on, here shown for the US owner-occupied market and for the income-tax preferences rather than for a property tax. The land-price framing is precisely the theory's prediction that the fixed factor absorbs the tax.

Limits a fair reader should note:

• It is a tax preference (subsidy), not a tax on land. The mechanism is the capitalization of the mortgage-interest/imputed-rent tax advantages; it shows capitalization runs in both directions (a subsidy capitalizes up, a tax capitalizes down), but it is not a direct study of a land value tax or of rental pass-through.
• Owner-occupied housing, not rentals. The incidence is read through owner transaction prices; the paper does not measure rents. (Its full-capitalization reading also implies tax preferences do little to reallocate real capital across structure types — a within-housing point, not a rental-incidence one.)[3]
• Model-plus-simulation, and a contested magnitude. The 13–17% figure is a simulation from an estimated user-cost model that assumes a single marginal taxpayer and abstracts from transaction costs; later "discrete-period" work (Harris 2013, the same secondary source) argues these assumptions overstate the price effect and that some reforms could even raise prices.[4] The direction — capitalization onto the owner via land value — is robust; the size is model-dependent.

Bears On

• Benefit (supports): Landlords cannot pass a land value tax on to tenants — shows tax variables are fully capitalized into residential land prices and borne by owners, the same asset-price incidence the claim relies on.
• Research: Palmon & Smith (1998) · Borge & Rattsø (2014) — property-tax capitalization analogues reaching the same full-capitalization conclusion.
• Research: Hilber (2017), capitalization synthesis — explains why the capitalized effect concentrates in land-scarce, supply-constrained metros, matching CGH's geographic variation.
• Concept: Tax capitalization.

See Also

• LVT improves housing affordability — the land-share logic behind the geographic variation
• Oates (1969), the founding capitalization study
• Mieszkowski (1972) — land as the factor that cannot escape the tax

Sources

1. Dennis R. Capozza, Richard K. Green & Patric H. Hendershott, "Taxes, Mortgage Borrowing, and Residential Land Prices," in Henry J. Aaron & William G. Gale (eds.), Economic Effects of Fundamental Tax Reform, Brookings Institution Press, 1996 — the published book chapter (venue verified; Brookings volume).
2. Working-paper version, "Taxes, Mortgage Borrowing and House (Residential Land) Prices" (Wisconsin CULER 96-06). Free full text: academia.edu/2820554 (title and authorship verified). Method and the "coefficients not different from one → taxes capitalized" account corroborated from co-author Richard Green's technical exposition of the CGH user-cost model (real-estate-and-urban.blogspot.com, fetched 2026-07-11). Note: academia.edu's auto-generated "FAQ" figures on that page are machine-produced and were NOT used.
3. Dennis R. Capozza, Richard K. Green & Patric H. Hendershott, "Taxes and House Prices" (Charles A. Dice Center Working Paper 97-16, Dec. 1997), SSRN abstract 48961 — the companion 63-metro (1970–1990) estimation; abstract fetched and verified (2026-07-11): "consistent with full capitalization of marginal income tax rates into house prices," with "substantial house price declines and wide geographic variation" under proposed reforms.
4. Benjamin H. Harris, "New Estimates of Tax Reform's Effect on Housing Prices," Urban-Brookings Tax Policy Center, Sept. 2013 — PDF — independent secondary source, fetched and verified (2026-07-11) — used for the 13–17% average price-decline summary attributed to CGH (1996) and for the later critique that transaction costs and non-marginal taxpayers reduce that estimate.