The Impact of Property Tax Incentives on Residential Investment (Baldomero-Quintana & Singh, 2026)

Summary

"The Impact of Property Tax Incentives on Residential Investment," by Luis Baldomero-Quintana and Divya Singh, published in International Tax and Public Finance (2026), studies a New York City property-tax-exemption reform to estimate how sensitive housing construction actually is to the terms of a recurring property tax. The paper is closely related to, but distinct from, the wiki's existing coverage of NYC's 1921–1931 new-housing tax exemption — this is a much more recent NYC reform (the 421-a-family exemption program's 2006/2008 eligibility tightening), studied with a modern short-run/long-run causal design.

The Reform and Method

NYC's tax exemption for new rental housing was announced in 2006 and its new, stricter eligibility requirements implemented in 2008. The authors exploit two distinct sources of variation:

• Short-run: the announcement-to-implementation gap. Because developers knew the exemption terms would tighten in 2008, they had a strong incentive to break ground before the deadline to lock in the old, more generous terms — a "grace period" dynamic. The authors use this gap to estimate the resulting excess housing starts.
• Long-run: a difference-in-differences design. Comparing census tracts where the new eligibility requirements were stringent against tracts where they were lenient, the authors estimate the reform's lasting effect on new construction once the temporary grace-period boom had passed.

Findings

• Short run: the tax reform induced 10,735 excess new rental-housing starts — equivalent to about 1% of the entire 2007 NYC rental housing stock — concentrated in the window before the stricter rules took effect.
• Long run: tracts subject to the strictest new eligibility requirements built 0.58 to 0.81 more new buildings and 22.25 to 22.75 more new rental units per census tract per year, relative to comparison tracts with lenient requirements. These effects are large relative to baseline activity — 0.3x to 0.4x the median new buildings per tract per year, and 2.4x to 2.5x the median new rental units per tract per year, over the 2003–15 sample period.

The authors draw two conclusions: first, that a property tax reform can raise residential investment in the long run even though a grace period first produces a temporary boom followed by a bust in construction timing; second, that local tax policy is "a key determinant of housing supply in cities with high density and land scarcity" — i.e., NYC's own tax-policy choices, not just its geography, shape how much housing gets built.

Relation to the Georgist Case

This is a within-NYC natural experiment in exactly the mechanism split-rate taxation is built on: changing the terms on which a recurring property tax applies to new residential construction produces a large, measurable construction response, both in the short-run timing of housing starts and in the long-run rate of new building. The paper doesn't test a land-value-only tax — it studies eligibility rules for a conventional building-tax exemption — but the underlying logic is the same one Oates & Schwab's Pittsburgh evidence and Plassmann & Tideman's Pennsylvania panel document: the tax treatment of new construction is a first-order determinant of how much gets built, in one of the most land-constrained major cities in the developed world.

The short-run finding is also a clean illustration of how sensitive developers are to anticipated future tax terms, not just current ones — directly relevant to the dynamics this wiki documents under speculative vacancy and the debate in the zoning-restrictions objection over how quickly supply responds to policy change once the incentive is in place.

Nuances and Limits

• Abstract-level scan. The full paper is paywalled behind a Springer bot-check this session (both WebFetch and curl returned a "Client Challenge" page); the findings above are drawn from the peer-reviewed abstract, retrieved via the Semantic Scholar API and verified verbatim against the abstract text returned there. No claim below the abstract level is made.
• Building-tax exemption, not a land tax. Like the paper above studying Murphy & Seegert's implicit land taxes, this paper's policy lever is a conventional property-tax exemption program (per-unit and per-building rules), not a land-value-only instrument — it is evidence for the general proposition that recurring property taxation shapes construction, and only indirectly for the land-specific case.
• Single city, single reform. The identification strategy is strong (a genuine announcement/implementation gap plus a tract-level difference-in-differences), but the result is one policy change in one dense, supply-constrained city; generalizing the magnitude to other markets should be done cautiously.

Bears On

• Benefit: Split-rate taxation increases construction — a modern, large-sample NYC natural experiment reinforcing the Pennsylvania evidence base with a different tax instrument and city.
• Research: Murphy & Seegert: Implicit Land Taxes — companion evidence that property-tax design (here exemption eligibility, there assessment weighting) shapes real construction and economic outcomes.
• Place: New York City

See Also

• NYC's 1921–1931 Tax Exemption for New Housing — the earlier NYC precedent for using tax exemptions to spur housing construction
• Split-Rate Taxation
• Oates & Schwab: Pittsburgh
• Plassmann & Tideman: Construction Response

Sources

1. Luis Baldomero-Quintana & Divya Singh (2026), "The Impact of Property Tax Incentives on Residential Investment," International Tax and Public Finance. DOI: 10.1007/s10797-026-09995-z — abstract retrieved via the Semantic Scholar API (api.semanticscholar.org) 2026-08-14 and verified verbatim against the returned abstract text, after both WebFetch and curl were blocked by a Springer bot-check page (SpringerLink "Client Challenge") on the article and PDF URLs directly. Used for the 2006/2008 reform mechanics, the short-run excess-starts estimate (10,735 units / 1% of 2007 rental stock), the difference-in-differences long-run estimates (0.58–0.81 buildings, 22.25–22.75 units per tract per year; 0.3x–0.4x and 2.4x–2.5x the sample median), and the authors' two stated conclusions. Scan depth: abstract only (B-claim); no claim below the abstract level is made.