|
|
- Info
Taxes, fees, and urban development
| Author: |
McFarlane, Alastair |
| Date: |
1999 |
| Periodical: |
Journal of Urban Economics |
| Abstract: |
Fiscal policy, when uncoordinated with urban planning, is an element that could bring about an inefficient urban structure. Illustrative of this position is the work of Gyourko and Voith [12], who find that urban sprawl in American cities may be attributable to the U.S. federal personal income tax system. A more optimistic view is that taxes can be structured in order to encourage socially optimal patterns of land development. For example, the development fee has been suggested as an alternative to growth controls [5] and exclusionary zoning [11]. In addition to development fees, both income and property taxes on real estate have been cited [2] as means of regulating development. One obstacle to advancing this type of research is the complexity of the comparative statics of real estate taxes. The difficulty is due to the interdependence of different decisions in the development process such as those concerning lot size, capital intensity, type of land use, timing, and whether to redevelop. Thus, most studies are restricted to addressing the effect of a tax on only one of the many aspects of development (see Anderson [1] and Turnbull [18] for a review of the literature on property taxes and the timing of development). However, to gain insight into the effects of a tax on urban structure in a dynamic economy, we need to investigate the effects of a tax on the timing of development together with a variable that describes the density of development. Turnbull [17], who examines the effects of a wide range of fiscal policies on both the timing of development and lot size, sets forth the most extensive conceptual analysis of fiscal policy, residential development, and urban structure. The model in his paper is such that development may occur at three different distances from the center of the city simultaneously, one of which is characterized by inward moving development. The implication for fiscal policy is that the effect of a tax on development depends upon the location of the site. This leads to a rich set of predictions concerning the fiscal policy and the growth and population density of a city. There are a few reasons for providing an alternative to Turnbull's analysis, one where the capital-to-land ratio is the variable of interest instead of lot size. First and most obvious, the effects of a tax on structural density may be a primary concern. Modeling the capital intensity of development is essential for understanding the effects of a tax change on the capital invested in building activity as well as the physical structure of the urban area. Although a number of authors [3, 4, 8, 14, 18] have already looked into the impact of different taxes on structural density, they examine neither a wide-range of taxes nor whether the effects vary by location. A second reason for treating the capital-to-land ratio as an endogenous variable is for expository purposes. When the per unit rent is independent of capital, the pattern of urban growth is simple: development moves outward from the center and there is a unique urban-rural boundary. While not as complex as the model of urban structure used by Turnbull, the model in this paper has the advantage of yielding less ambiguous results. Moreover, with the simpler growth pattern, it is possible to use the hurdle rent approach for explaining the comparative static predictions. The third and final contribution of this paper is a discussion of the effects of different types of development fees. In most conceptual analyses, fees are modeled as lump-sum taxes imposed upon developers. However, in practice, fees vary by the type of development, number of rooms, land value, acreage, square footage, development density, and service area. Thus, whether developers react differently to different kinds of impact fees should be a concern.
In the next and second section, a model of urban growth with durable housing is presented. In the third section, the model is used to develop comparative static predictions concerning the effect of fees and taxes on developer behavior. In the fourth and final section, a summary of the results is presented.
|
|
