Is the Property Tax a Benefit Tax? The Case of Rental Housing

Summary

"Is the Property Tax a Benefit Tax? The Case of Rental Housing" is a 1994 article by Robert Carroll and John Yinger, published in the National Tax Journal, vol. 47, no. 2 (June 1994), pp. 295–316 (DOI: 10.1086/NTJ41789069). Yinger, then at Syracuse University's Maxwell School, is one of the property-tax-incidence literature's most active empirical contributors and a co-author of the standard reference work on capitalization (Yinger, Bloom, Börsch-Supan & Ladd, Property Taxes and House Values, 1988), which is itself the benchmark study cited in Palmon & Smith (1998). This paper matters specifically for the gap the other sources supporting this outcome do not close: Mieszkowski (1972) and Zodrow (2001) are general-equilibrium theory, not direct measurement, and Palmon & Smith (1998) measures capitalization into owner-occupied home sale prices, not rents paid by tenants. Carroll & Yinger instead take the incidence question directly into the rental market: using data on 147 towns and cities in the Boston metropolitan area in 1980, they estimate how rents actually respond to differences in local property-tax rates paid by landlords, holding housing and neighborhood characteristics constant. Published in the field's leading U.S. policy-oriented tax journal, this is a mainstream, peer-reviewed empirical test of tenant-side incidence — not a Georgist or advocacy publication — and it is one of relatively few studies of its kind to examine actual market rents rather than owner-occupied values or theoretical models.

The Core Argument and Findings

The paper is framed as a test of the "benefit view" of the property tax (associated with Hamilton's benefit-tax view): the hypothesis that a property tax funding local public services functions as a payment for those services rather than as a distorting tax, so that a tenant who is fully "mobile" (able to move costlessly between jurisdictions) should, in equilibrium, pay higher rent only insofar as the additional tax buys better local services they value. Carroll & Yinger set up the most favorable possible case for that hypothesis — fully mobile tenants — and test it against actual Boston-area rental data.

Using a hedonic rent equation with controls for housing unit characteristics, neighborhood quality, and local public-service levels, they estimate the direct effect of a local property-tax-rate increase on gross rent, isolated from any offsetting change in services the tax revenue buys. Their central empirical result: a $1.00 increase in property taxes is associated with a rent increase of only about $0.15, on average — implying landlords absorb roughly 85 percent of a property-tax increase directly, with only a small share showing up as higher rent. In the authors' own words: "In our sample, a $1.00 property tax increase results in a rent increase of only about $0.15, on average, even with infinitely elastic housing supply" (p. 310); under their tenant-mobility scenario "the net burden on tenants from a $1.00 increase in property taxes is $0.156, on average, with a minimum of $0.104 and a maximum of $0.253" across communities, leaving "the average burden on landlords... $0.844 for every $1 increase in property taxes" (pp. 308–309). The pattern thus holds broadly across their sample — the landlord's share of the tax differential ranges from roughly 75 to 90 percent across the towns and cities studied — rather than being an artifact of a few outlying jurisdictions.

Carroll & Yinger then ask whether the indirect channel — tenants valuing the additional public services the tax revenue funds — closes the gap, which is the condition the benefit-tax hypothesis actually requires (full compensation via services, not via rent). They find it does not: even accounting for tenants' estimated willingness to pay for the resulting service improvements, the indirect, service-side benefit covers only about $0.77 of each $1.00 of additional property tax, on average, in their Boston-area data — as the abstract puts it, "on average, indirect shifting covers only about $0.77 of a $1.00 increase in property taxes" (p. 295). Combining the direct (rent) and indirect (service-value) channels, the authors conclude that renters do not bear the full burden of a property-tax increase, and the property tax on rental housing is not, in practice, a benefit tax — landlords bear a substantial net burden even after crediting tenants with the value of services financed by the higher tax: "Regardless of one's assumptions about the housing price elasticities, therefore, our analysis of two necessary conditions reveals that the property tax is far from being a benefit tax both on average and in every community" (p. 309).

Relation to the Georgist Case

This paper supplies exactly the kind of evidence the outcome Landlords cannot pass a land value tax on to tenants had been missing: a direct empirical estimate of tax incidence in an actual rental market, rather than a theoretical model (Mieszkowski, Zodrow) or a study of owner-occupied sales (Palmon & Smith). The finding — that the great majority of a property-tax increase is absorbed by landlords rather than passed through as higher rent, and that this holds even in the "most favorable" case of mobile tenants where economic theory would predict the most shifting — is squarely supportive of the claim that landowners, not tenants, bear the incidence of a tax that falls on a fixed local asset.

Two scope qualifications matter for how strongly this paper should be read as evidence for LVT specifically, and both should be stated plainly. First, Carroll & Yinger study the conventional U.S. property tax — land and structures/improvements assessed and taxed together — not a land-value-only tax. Because a general property tax also taxes the building, which is not fixed in supply, standard incidence theory (see Mieszkowski) predicts more of it, not less, should be shiftable than a pure LVT would be; the low measured pass-through here is therefore, if anything, a conservative (harder) test that a land-only tax would be expected to pass even more decisively. Second, this is a single metropolitan area (Boston) at a single point in time (1980) — a real limitation on how far the specific $0.15-on-the-dollar and 85-percent figures generalize to other cities, tax structures, or eras; the paper should be cited for its directional finding and its rejection of full/near-full shifting, not as establishing a universal pass-through rate.

Nuances and Limits

• General property tax, not LVT. As above, the tax studied is the ordinary property tax on land plus improvements. The paper does not isolate a land-only component, so it cannot by itself distinguish "landlords absorb the tax because land is fixed in supply" from "landlords absorb the tax because of some other rigidity in this particular rental market" — though the fixed-land-supply mechanism is the one general-equilibrium theory (Mieszkowski, Zodrow) predicts should dominate for the land share specifically.
• Single metro area, single year. The 147-town Boston-area, 1980 cross-section is a real strength for internal identification (many jurisdictions with varying tax rates, comparable regional housing market) but a limit on external validity; rental markets with different vacancy rates, rent regulation, or supply elasticities elsewhere could show different pass-through.
• Benefit-tax framing, not a pure incidence study. Because the paper is designed to test the benefit view, its "indirect shifting" estimate (services covering $0.77 of $1.00) is itself a modeled willingness-to-pay figure, not a directly observed price — a further layer of estimation uncertainty beyond the direct rent-pass-through regression.
• Consistent with, but does not on its own prove, the LVT-specific claim. The paper is strong, direct evidence that rental markets in practice do not fully shift a property-tax increase onto tenants; it does not purport to model a hypothetical land-value-only tax, so a reader should describe it as strongly supportive analogous evidence rather than as a direct test of LVT incidence.
• Contrast worth noting honestly. Not all rental-market incidence evidence agrees on magnitude. Lyndsey Rolheiser's study of commercial office rents in Massachusetts (MIT dissertation, 2017/2019) finds substantially higher pass-through — on the order of 80–90% of a property-tax change absorbed into commercial rents — and Max Löffler & Sebastian Siegloch's study of German municipal property-tax (Grundsteuer) reforms finds much higher pass-through to residential rents. Their 2021 working paper reports that "in the short run—i.e., up to two years after the policy change—around one third of the additional property tax due is passed-through to consumer prices," while "in the medium to long-run—i.e., after three years—the point estimates imply full path-through of tax payments"; the revised July 2024 version, using 5,500 municipal tax changes, states the headline estimate as "83 percent of the tax burden is passed through to rental prices." Consistent with the supply-elasticity mechanism, they find "the pass-through is lower when housing supply is inelastic," and their welfare simulations conclude the German property tax is regressive because of this high pass-through — it "could be progressive... if there was zero pass-through from landlords to renters." On the assessment base, the paper is explicit that the tax is not land-value-only: "The German property tax (Grundsteuer B) is a one-rate tax that applies to the land and built structures," levied on assessed unit values (Einheitswert) frozen at 1964 levels with no regular reassessment. (Löffler & Siegloch also situate Carroll & Yinger within the earlier U.S. literature, noting that the pre-existing studies "estimate that between 0–115 percent of the tax is shifted onto renters.") These studies differ from Carroll & Yinger in tax base (Germany's Grundsteuer and Massachusetts commercial property taxes are not land-value-only), sector (commercial office space vs. general rental housing), country, and era, so they should not be read as directly overturning Carroll & Yinger's Boston finding — but they are a reminder that residential-rental incidence is not a fully settled empirical question across all contexts, and that pass-through appears to vary with local housing-supply elasticity and market structure. A balanced treatment of this outcome should acknowledge this variation rather than presenting Carroll & Yinger's low pass-through as the sole empirical finding in the literature.

Bears On

• Outcome: Landlords cannot pass a land value tax on to tenants — the paper's central contribution: direct empirical evidence, from an actual rental market, that landlords absorb the large majority of a property-tax increase rather than passing it to tenants, filling the gap left by theory-only and owner-occupied-sales evidence.
• Research: Mieszkowski (1972), the "new view" of property tax incidence — supplies the general-equilibrium theoretical prediction (fixed-supply land borne by owners) that this paper's rental-market data are broadly consistent with.
• Research: Zodrow (2001), "A Room with Three Views" — this paper is an empirical test bearing on the benefit view Zodrow surveys as one of the three rival incidence traditions.
• Research: Hamilton's benefit-tax view — Carroll & Yinger directly test, and substantially reject, the benefit-tax hypothesis for rental housing in their Boston data, even under the most favorable (fully mobile tenant) assumptions.
• Research: Palmon & Smith (1998), property tax capitalization — the natural complement: Palmon & Smith measure capitalization into owner-occupied sale prices, while this paper measures the analogous incidence question directly in the rental market.
• Concept: Tax Capitalization — the low rent pass-through is the rental-market expression of the same capitalization mechanism documented for home sales elsewhere in this literature.

See Also

• Löffler & Siegloch — German property-tax pass-through
• Landlords cannot pass a land value tax on to tenants
• Mieszkowski (1972), the "new view" of property tax incidence
• Zodrow (2001), "A Room with Three Views"
• Palmon & Smith (1998), property tax capitalization
• Hamilton's benefit-tax view
• Tax Capitalization

Sources

1. Robert Carroll & John Yinger (1994), "Is the Property Tax a Benefit Tax? The Case of Rental Housing," National Tax Journal 47(2): 295–316. DOI: 10.1086/NTJ41789069 (paywalled via University of Chicago Press) · full text via the NTA's open NTJ archive, Internet Archive snapshot — primary source for the paper's methodology (147 Boston-area towns/cities, 1980), the $0.15-per-$1.00 direct rent pass-through estimate, the ~85% landlord burden share, and the $0.77-per-$1.00 indirect service-value estimate. Full text read directly on 2026-07-10 via the archived copy of the National Tax Association's open NTJ archive; all quantitative findings and quotations above verified against the original with page-level citations (abstract p. 295; tenant/landlord burden estimates pp. 308–309; benefit-tax rejection p. 309; conclusion pp. 310–313).
2. John M. Yinger, "PAI735/ECN635 – Notes on the Incidence of the Property Tax," Syracuse University (Maxwell School) course materials. Yinger's site — used as a secondary summary, by one of the paper's own authors, of the Boston-area data, the $0.15/$1.00 finding, and the paper's relation to the benefit-tax and traditional views.
3. John Yinger, Howard S. Bloom, Axel Börsch-Supan & Helen F. Ladd (1988), Property Taxes and House Values: The Theory and Estimation of Intrajurisdictional Property Tax Capitalization, Academic Press — the companion capitalization study by an overlapping author team, also cited in Palmon & Smith (1998), used here for context on Yinger's broader body of incidence work.
4. Athiphat Muthitacharoen & George R. Zodrow (2012), "Revisiting the Excise Tax Effects of the Property Tax," National Tax Journal — used as a secondary cross-check that situates Carroll & Yinger's low forward-shifting estimate among other quantitative incidence studies.
5. Lyndsey Rolheiser, "Commercial Property Tax Incidence: Evidence from the Rental Market," MIT PhD dissertation (2017); summarized in MIT Center for Real Estate, "Can Landlords Really Pass on Higher Property Taxes to Tenants?" MIT CRE — used for the contrasting commercial-office-rent pass-through estimate (80–90%) noted in "Nuances and Limits."
6. Max Löffler & Sebastian Siegloch, "Welfare Effects of Property Taxation," CESifo Working Paper No. 8952 / IZA Discussion Paper No. 14195 (March 2021); revised as ECONtribute Discussion Paper No. 331 (July 2024). IZA · ZEW · ECONtribute 2024 revision — used for the contrasting German Grundsteuer pass-through-to-rents finding noted in "Nuances and Limits." Both the 2021 IZA/ZEW text and the 2024 revision were read directly: the 2021 abstract states "higher taxes are fully passed on to rental prices after three years," the 2024 abstract states "83 percent of the tax burden is passed through to rental prices," and Section 3.1 of both versions confirms the Grundsteuer B applies to "the land and built structures" on 1964-frozen assessed values — all quotations in "Nuances and Limits" verified against these primary texts.