Capitalization of Property Taxes in Norway

Summary

"Capitalization of Property Taxes in Norway" by Lars-Erik Borge and Jørn Rattsø (Norwegian University of Science and Technology, Trondheim) was published in Public Finance Review 42(5), 2014, pp. 635–661 (first published online 2013).[1] It is a well-identified capitalization study whose setting is close to a natural experiment: Norwegian municipalities choose whether to levy a residential property tax, and many do not, generating exogenous cross-municipal variation in tax exposure that the authors exploit with instrumental variables.[1]

Using a rich data set of housing transactions and characteristics for 1997–1999 combined with municipality-level data on property taxation, local services, and community characteristics, the authors ask whether property-tax differences show up as differences in house prices. Their answer, verified against the full text (2026-07-11):[1]

"The results indicate that housing prices respond to property taxation and with full capitalization at realistic discount rates."[1]

The paper matters for the wiki's landlords-cannot-pass-LVT-to-tenants claim because full capitalization is exactly the mechanism the incidence argument relies on: the authors state plainly that "full capitalization implies that current owners bear the entire burden of expected tax liabilities, whereas partial capitalization suggests that some of the burden is passed on to future owners."[1] When a tax is fully capitalized into the asset price, the person who owns the asset at the moment of the change absorbs the whole discounted stream — there is no channel to shift it forward.

Core Findings

• Full capitalization at realistic discount rates. Their estimates imply that raising the effective property-tax rate by 0.1 percentage point reduces housing prices by nearly 3.5 percent; for the sample's average house (about NOK 1 million) that is roughly a NOK 35,000 price drop against a NOK 1,000 annual tax increase — consistent with full capitalization at a discount rate of about 2.9 percent.[1] A separate dummy-variable formulation (houses are about NOK 60,000 cheaper in property-taxing municipalities, against ~NOK 1,350 annual tax) implies full capitalization at about a 2.3 percent discount rate; the two formulations agree.[1] These are real (inflation-adjusted) discount rates, in the same range as Norwegian real deposit and loan rates over the period.
• Identification via IV. Cross-sectional capitalization studies are dogged by endogeneity — tax rates correlate with unobserved service quality and community characteristics that also move house prices. Borge & Rattsø instrument for property taxation (using features of the local political system, e.g. the party composition of the council) and conclude: "the instrument estimation support our interpretation of full property tax capitalization in this data set."[1]
• Situates itself with the US evidence. The authors read their result as consistent with recent US capitalization work, singling out Palmon & Smith (1998), and note that the older US/Canadian literature (surveyed by Yinger et al. 1988) mostly found partial capitalization in the 15–65 percent range, attributing the shortfall chiefly to expectations that tax differentials would not persist.[1] Their finding that Norwegian tax differences are "expected to persist" is what pushes their estimate to full capitalization.

What it Supports — and its Limits

Supports the capitalization pillar of the incidence argument. This is a second, non-US, quasi-experimental confirmation — alongside Palmon & Smith and the Danish land-tax study — that tax differentials are absorbed by the asset price, i.e. by the owner, rather than shifted forward. It strengthens the tax-capitalization mechanism that underwrites the claim that a land value tax cannot be passed to tenants.

Limits a fair reader should note:

• It is the property tax, not a pure land tax. The Norwegian municipal property tax falls on the whole property (structure plus site), so this is capitalization of a building-inclusive tax, like Palmon & Smith and unlike the Danish land-value tax. Full capitalization of a building-inclusive tax is a stronger result for the land case than the theory strictly requires, but it is still not a direct test of a land-only levy.
• Owner-occupied prices, not measured rents. The study measures capitalization into transaction prices, not pass-through into rents; it speaks to incidence through the asset-price channel, which is where the theory locates the burden, but it is not a rental-market measurement like Carroll & Yinger or Löffler & Siegloch.
• Discount-rate dependence. As in all capitalization work, "full" is a statement relative to a chosen discount rate; the authors' realistic 2.3–2.9 percent real rates are defensible but the mapping from coefficient to capitalization share is sensitive to that choice.

Bears On

• Benefit (supports): Landlords cannot pass a land value tax on to tenants — supplies a non-US, IV-identified confirmation that property-tax differentials are fully capitalized into owners' asset prices, closing the forward-shifting channel.
• Research: Palmon & Smith (1998) — the US study Borge & Rattsø benchmark against; both cannot reject full capitalization at realistic discount rates.
• Research: Danish Economic Councils (2018), land-tax capitalization — the land-tax-specific analogue; Borge & Rattsø is the property-tax analogue in a comparable Nordic setting.
• Concept: Tax capitalization — a textbook instance of the mechanism.

See Also

• Oates (1969), the founding capitalization study
• Hilber (2017), capitalization synthesis — surveys why capitalization is fuller where housing supply is more inelastic
• Capozza, Green & Hendershott (1996), residential land prices
• Mieszkowski (1972) · Zodrow (2001) — the incidence theory

Sources

1. Lars-Erik Borge & Jørn Rattsø, "Capitalization of Property Taxes in Norway," Public Finance Review 42(5), 2014, pp. 635–661 (published online 2013). DOI 10.1177/1091142113489845. Free full-text PDF (author-affiliated copy, gwern archive): gwern.net/doc/economics/georgism/2013-borge.pdf — fetched and read in full (2026-07-11) — used for and verified against all findings on this page: the abstract's full-capitalization conclusion, the owner-bears-the-burden definition, the ~3.5%-per-0.1pp and NOK-60,000 dummy estimates, the 2.3–2.9% implied discount rates, the IV identification, and the comparison with Palmon & Smith and the Yinger et al. survey.