LVT can replace capital taxes without efficiency loss

At a glance — Because land is fixed in supply and bears no deadweight loss while capital taxes deter investment at the margin, a revenue-neutral swap from capital taxes to a land tax raises welfare — supported by theory and calibrated general-equilibrium models. Evidence: Strong for the core no-deadweight-loss result (theory + calibrated general-equilibrium models); two qualifications now carried as counter-evidence · 8 supporting sources · 2 challenging Strongest support: Bonnet et al. (2021) — taxing land dominates taxing capital and can substitute for capital taxes without the efficiency cost. Strongest counter: Feldstein (1977) — in a growth model where land and capital compete as stores of wealth, a tax on pure rent can be "at least partly shifted," so even a land tax is not unambiguously neutral.

The Claim

Replacing taxes on capital with a tax on land improves economic efficiency with no offsetting loss. Because land is fixed in supply it bears zero deadweight loss, whereas taxing capital reduces investment at the margin. A revenue-neutral swap therefore raises total welfare.

The Evidence

This rests on the oldest result in the field: a tax on a factor in perfectly inelastic supply causes no change in quantity, hence no deadweight loss — a point on which economists from Henry George to Milton Friedman (who called LVT the "least bad tax") agree.

The Evidence in Detail

The institutional and theoretical record behind the ranking. Johansson et al. (2008) is the OECD's famous "tax and growth" hierarchy — recurrent taxes on immovable property least harmful to growth, corporate income taxes most — and Arnold et al. (2011) is its peer-reviewed Economic Journal version: revenue-neutral shifts toward recurrent property taxation raise long-run GDP per capita across a 21-country OECD panel. The Mirrlees Review (2011) turns the ranking into institutional advice, concluding land value should be taxed and business rates replaced with an LVT. Brueckner (1986) supplies the formal split-rate theory: shifting tax off improvements onto land raises the capital-to-land ratio — the mechanism behind the efficiency claim.

Counter-Evidence and Limits

The core claim — that a tax on land's fixed supply changes no production decision while capital taxes deter investment at the margin — is among the most secure in public finance. Two qualifications keep it from being asserted as a literal, unconditional zero.

• A land tax may not be perfectly neutral (the portfolio-shifting challenge). Feldstein (1977) shows that in a life-cycle growth model where land and capital are competing stores of retirement wealth, a tax on pure land rent is "at least partly shifted" — the price of land can even rise — because taxing land pushes household savings toward produced capital, changing the capital stock. If so, the land tax is not the perfectly inert instrument the simplest static argument assumes. Crucially, this does not reverse the efficiency comparison with the capital taxes it replaces; and later work (Calvo, Kotlikoff & Rodriguez 1979; Fane 1984) restores the classical no-shifting result under bequest and full-compensation assumptions, so the challenge sharpens the case rather than overturning it.
• The assessment basis matters (the Bentick–Mills timing distortion). Bentick (1979) and Mills (1981) show that a land tax assessed on market value — which capitalizes the highest-and-best-use option — can distort the timing of development, biasing owners toward quick-yield uses. That is a genuine efficiency cost, but Tideman's 1982 rebuttal locates it in the assessment basis (taxing development-option value) rather than in land taxation itself: a tax on current-use site value avoids it. The efficiency claim is therefore strongest for a correctly-assessed site-value tax.

Neither result reverses the ranking of a land tax above the capital taxes it would replace. Both mean the headline should be read as "with far less efficiency loss than the alternative, and none under the idealised assumptions" — not as a literal zero under every model and every assessment practice.

Strength of Evidence

Strong — grounded in well-established theory and confirmed by independent calibrated models, with the two qualifications above (imperfect neutrality under portfolio-shifting; assessment-basis timing effects) bounding, not overturning, the result.

See Also

• Taxing land and rents increases productivity — the stronger growth claim built on this page's well-supported core, graded honestly
• Deadweight Loss · ATCOR · Land Value Tax
• Taxation of Economic Rents (Schwerhoff, Edenhofer & Fleurbaey 2020) — the peer-reviewed survey backing the non-distortion case
• Feldstein (1977): the portfolio-shifting incidence challenge — the canonical caveat that even a pure land tax may not be perfectly neutral
• The Bentick–Mills timing critique — why market-value assessment can distort development timing, and Tideman's rebuttal

Sources

1. Bonnet, Chapelle, Trannoy & Wasmer (2021), European Economic Review — used for the result that shifting taxes onto land (away from capital) is feasible and non-distortionary given land's large, taxable share of wealth. wiki summary · PDF
2. Schwerhoff, Edenhofer & Fleurbaey (2022), "Equity and Efficiency Effects of Land Value Taxation," IMF Working Paper — used for the optimal-tax finding that an LVT can substitute for distortionary capital taxes without efficiency loss. PDF