The Bentick–Mills Timing Critique (and the Tideman Rebuttal)

Summary

This page carries the development-timing critique of land-value-tax neutrality — the standard theoretical counter-evidence in the incidence and land-use literature, presented per the wiki's steelman rule.

• Bentick (1979), "The Impact of Taxation and Valuation Practices on the Timing and Efficiency of Land Use" (Journal of Political Economy 87(4), pp. 859–868): a land tax assessed on current market value (which capitalizes future development options) is not neutral — it biases landowners toward projects with short gestation periods, accelerating and distorting development timing at real resource cost.[1]
• Mills (1981), "The Non-Neutrality of Land Value Taxation" (National Tax Journal 34(1), pp. 125–129): the companion result — with endogenous land-market adjustments, a tax on land value is non-neutral toward development decisions.[2]

For the sprawl outcome the implication is pointed: a market-value-assessed land tax could hasten conversion at the urban fringe — the opposite of the anti-sprawl claim — which is why this page is wired as challenged_by there.

The Rebuttal — the Dispute Is About the Assessment Basis

Tideman (1982), "A Tax on Land Value Is Neutral" (National Tax Journal), replies that the Bentick–Mills distortion comes from taxing the capitalized value of development options, not from taxing land as such: a tax assessed on pre-development site value (what the land would be worth absent the owner's own development plans) restores neutrality.[3] Later theoretical work in the same exchange — Richard Arnott's "Neutral Property Taxation" (2005) — develops the same distinction.[4] Arnott reviews Shoup (1970), Skouras (1978), Bentick (1979) and Mills (1981) — "the revisionists" — who showed that taxing residual site value (property value minus the depreciated cost of the structure) is non-neutral, in particular discouraging density, and confirms with Tideman (1982) that neutrality is achieved only when the base is raw site value — "what the site would be worth if there were no structure on it," a value "unaffected by the developer's decisions" (verified verbatim against the Boston College working-paper text).

The wiki's honest framing: both sides are right about different taxes. Real assessment regimes vary; where valuation includes development option value, the Bentick–Mills distortion is live (and, note, its timing bias reinforces the pro-density finding in Banzhaf & Lavery while complicating its welfare interpretation — their own timing-effect caveat). Where site value is assessed as the classical theory intends, neutrality survives. The practical question becomes an assessment-design question.

Bears On

• Outcome (challenges): LVT reduces urban sprawl — the timing channel can cut the other way
• Concept: Deadweight Loss — the neutrality claim this critique qualifies
• Objection (adjacent): Land cannot be assessed — where the dispute lands in practice

See Also

• Brueckner (1986) — the neutrality-side formal treatment
• Nicolaus Tideman — author of the rebuttal
• Feldstein (1977), The Surprising Incidence of a Tax on Pure Rent — a distinct theoretical challenge to land-tax neutrality/incidence, working through a savings-portfolio channel rather than the development-timing/assessment channel treated here

Sources

1. Brian L. Bentick, "The Impact of Taxation and Valuation Practices on the Timing and Efficiency of Land Use," Journal of Political Economy 87(4), 1979, pp. 859–868. DOI — used for the timing non-neutrality result (C/E-claim).
2. David E. Mills, "The Non-Neutrality of Land Value Taxation," National Tax Journal 34(1), 1981, pp. 125–129. Publisher — used for the companion non-neutrality result (C/E-claim).
3. T. Nicolaus Tideman, "A Tax on Land Value Is Neutral," National Tax Journal,
4. PDF (cooperative-individualism) — used for the assessment-basis rebuttal (C-claim).
5. Richard Arnott, "Neutral Property Taxation," Journal of Public Economic Theory 7(1), 2005, pp. 27–50 (originally Boston College Working Paper 411, September 1998). Free full text (BC WP 411) — used for the residual-vs-raw site-value neutrality distinction and its explicit engagement with Shoup (1970), Skouras (1978), Bentick (1979), Mills (1981) and Tideman (1982); verified verbatim this session.