Working Paper: NBER ID: w8913
Authors: Richard Arnott; Petia Petrova
Abstract: Consider an atomistic developer who decides when and at what density to develop his land, under a property tax system characterized by three time-invariant tax rates: the tax rate on pre-development land value, the tax rate on post-development residual site value, and the tax rate on structure. Arnott (2002) identified the subset of property value tax systems which are neutral. This paper investigates the relative efficiency of four idealized, non-neutral property value tax systems (Canadian property tax system, simple property tax system, residual site value tax system, and differentiated property tax system) under the assumption of a constant rental growth rate.
Keywords: No keywords provided
JEL Codes: H2
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Canadian property tax system (exempting land prior to development) (H13) | more efficient taxation outcome (H21) |
| taxing postdevelopment residual site value and structure value at the same rate (H29) | maximizes tax revenue without distorting development decisions (H21) |
| property tax rate set below the growth rate of rents (R38) | significant inefficiencies (D61) |
| one percent property tax on property value (H82) | marginal deadweight loss of two dollars (H21) |
| some forms of property taxation (R51) | very inefficient (D61) |
| shift towards less distortionary tax systems (H21) | reduced inefficiencies (D61) |