Keywords

JEL Classifications

Property taxation of both residential and non-residential land and structures – or ‘real property’ – is the most common form of wealth taxation worldwide, and is often the revenue instrument of choice for local governments. For example, in the United States property taxes account for over 70 per cent of local own-source tax revenues. Property tax liability is calculated as the product of the statutory rate and the assessed tax base, subject to a variety of adjustments, such as partial exemptions for primary residences and ‘circuit breakers’ designed to reduce tax burdens for certain groups, especially relatively poor elderly homeowners. Although vagaries in the assessment process have long been a source of inequity in the administration of the tax, recent advances in computer-based assessment practices have mitigated this problem. More recently, rapid growth in residential home values and the concomitant increase in property tax burdens and the share of total local taxes paid by homeowners have led to increasing dissatisfaction with the tax in some regions, culminating in the passage of numerous property tax limitation measures. In addition, concerns about the equity implications of financing primary and secondary public education with the property tax when the tax base, including non-residential property, is unequally distributed across school jurisdictions have led to reduced reliance on the tax as well as equalization mechanisms that redistribute property tax revenues across school districts (Anderson 1994).

The Incidence Debate: The Three Views of the Property Tax

The academic literature on the property tax – as reviewed by Mieszkowski and Zodrow (1989), Ladd (1998), Ross and Yinger (2000) and Netzer (2001) – has focused on the incidence and effects on the allocation of resources of the residential property tax. There is general agreement that the land component of the property tax is capitalized into land values, is borne by landowners at the time of the imposition of the tax, and – since land is immobile – does not distort the allocation of resources. Indeed, the efficiency advantages of taxing land values, coupled with a belief that most increases in land values reflect the benefits of public services, have led some observers, most prominently Henry George (1879), to advocate replacing property taxes with taxes on land values.

By comparison, the incidence and economic effects of property taxation of the capital or structures component of real property are among the most contentious issues in state and local public finance. Three views dominate the debate. The ‘traditional’ view dates back to Simon (1943) and Netzer (1966), who focused on the partial equilibrium effects of increasing the tax in a local housing market. From this perspective, one can make the standard ‘open economy’ assumption that the national return to capital is fixed. This in turn implies that local capital bears none of the local property tax, as capital in the long run migrates from the jurisdiction until the local after-tax return to capital equals the national value. The burden of the tax is thus borne by local factors and/or consumers, with the traditional view holding that the entire burden is borne by local housing consumers. The traditional view thus implies that the property tax inefficiently reduces the size of the local housing stock and that its burden is borne in proportion to housing consumption – and is thus somewhat regressive with respect to annual income but, more importantly, roughly proportional with respect to lifetime income.

A second popular theory is the ‘benefit tax’ view, developed by Hamilton (1975, 1976); Fischel (2001a, b) provides a recent discussion. This view is an extension of the renowned Tiebout (1956) model, which argues that consumer mobility (‘voting with the feet’) and inter–jurisdictional competition in the provision of local public services can ensure efficiency of resource allocation in the local public sector. Although Tiebout assumed the existence of benefit/head taxes, Hamilton extended the analysis by deriving conditions under which the property tax can be converted into the head tax assumed by Tiebout.

Specifically, following Tiebout, Hamilton assumes that individuals sort into local jurisdictions according to their demands for local public services, and that there are enough local tax-expenditure packages to accommodate all tastes. In addition, Hamilton (1975) assumes that local jurisdictions are homogeneous with respect to house values, and that there are enough jurisdictions to accommodate all desired housing and government service/tax packages. Finally, Hamilton assumes the existence of binding zoning constraints that established a minimum house value for each community. Under these circumstances, individuals are precluded from purchasing homes with a value below the minimum, and would never elect to pay taxes in excess of benefits received by purchasing a home with a value greater than the minimum; all individuals in a given community thus pay exactly the same property tax, which becomes a benefit tax.

Hamilton (1976) extends this model to the more realistic case in which house values within a community are heterogeneous. In this case, Hamilton assumes all communities are fully developed, effectively precluding any tax-induced changes in the housing stock, and that alternative communities which are homogeneous with respect to both demands for public services and housing are available. Under these circumstances, Hamilton shows that ‘perfect capitalization’ converts the property tax into a benefit tax, as a relatively expensive home sells at a discount equal to its ‘fiscal differential’ or the present value of all future taxes in excess of benefits received, while a relatively inexpensive home sells at a premium reflecting its fiscal differential, the present value of all future benefits in excess of future taxes. The implications of the benefit tax view are striking, as it implies that the property tax is effectively a non-distortionary user charge that has no direct effects on income distribution.

Finally, the ‘capital tax’ view (or ‘new’ view) of the property tax, developed by Mieszkowski (1972), subsequently extended by Zodrow and Mieszkowski (1983, 1986b), and reviewed in Zodrow (2001a, b), argues that the property tax is a distortionary tax on the local use of capital that results in a misallocation of the national capital stock across local jurisdictions. Mieszkowski (1972) stresses that earlier partial equilibrium analyses ignored the fact that the property tax is used by virtually all local jurisdictions and applies to a large fraction of the capital stock, including most non-residential capital. Adapting the Harberger (1962) general equilibrium model of tax incidence, he models the economy as having a fixed national capital stock and two types of local jurisdictions, characterized by ‘high’ tax rates and ‘low’ tax rates. In this context, Mieszkowski shows that property tax rates that exceed the national average drive capital out of high-tax jurisdictions into low-tax jurisdictions, with opposing effects occurring in relatively low tax jurisdictions. Property tax differentials thus result in an inefficient allocation of capital across jurisdictions. In addition, concern about the extent to which use of the property tax may drive capital out of a jurisdiction creates a tendency for local governments to under-provide public services (Zodrow and Mieszkowski 1986b; Wilson 1986). In terms of incidence, the ‘average burden’ of all of the property taxes imposed across the nation – known as the ‘profits tax’ effect of the tax – is borne by capital owners generally, implying that the tax is relatively progressive (with respect to annual income). In addition, Mieszkowski stresses that property tax differentials about the national average result in ‘excise tax effects’ in the form of housing and commodity price increases and wage and land price declines in relatively high-tax jurisdictions, with offsetting effects in relatively low-tax jurisdictions.

Differentiating Among the Three Views

Much of the subsequent literature has focused on reconciling or differentiating among these three views, and the issue of which view most accurately describes the effects of the property tax is still contentious. Matters are simplified somewhat because the traditional view has been shown to be a special case of the capital tax view. Specifically, the traditional view can be interpreted as a partial equilibrium analysis that focuses exclusively on the excise tax effects of the capital tax view, while neglecting its general equilibrium profits tax effects. Moreover, the traditional view that these excise tax effects are fully reflected in higher housing prices is accurate only under special circumstances; more generally, excise tax effects are borne in some combination by housing consumers and the owners of labour and land in the taxing jurisdiction (Wildasin 1986). In addition, the profits tax effect still obtains when one takes a general equilibrium perspective of the use of the property tax by a single small jurisdiction facing a perfectly elastic supply of capital. Specifically, although the capital outflow caused by an increase in the property tax by a small local jurisdiction depresses the overall return to capital only very slightly, this reduction affects a large capital stock; under certain circumstances, the overall reduction in national capital income precisely equals the revenue raised by the taxing jurisdiction, yielding the profits tax result (Zodrow and Mieszkowski 1983; Brown 1924; Bradford 1978). At the same time, the burden of a property tax increase in a single jurisdiction is borne entirely by local residents as higher prices or lower factor returns (with offsetting effects in all other jurisdictions). A critical implication is that even under the capital tax view there is a close link between local public services and the burden of the property tax, as the cost of financing local expenditures largely falls on local factor owners and consumers; thus, this interpretation of the capital tax view clearly has a strong benefit view flavour as local residents ‘pay for what they get’ in public services.

Nevertheless, the debate between proponents of the benefit tax view and the capital tax view is still far from resolved. The original Mieszkowski (1972) derivation, based on a model of national tax incidence, has been criticized for ignoring many of the features of local government service provision stressed by Tiebout and Hamilton. However, Zodrow and Mieszkowski (1986a) present an expanded derivation of the capital tax view that includes most of these aspects, including interjurisdictional competition, varying tastes for local public services, individuals sorting into differing communities according to tastes for local public services, and a simple form of land use zoning. They conclude that these factors thus do not distinguish between the two views; instead, the key factor in determining the incidence of the property tax is whether housing consumption can vary in response to the imposition of the tax. Moreover, although zoning requirements are pervasive, take a wide variety of forms, and can have a significant impact on property values (Fischel 1992), these facts do not demonstrate that zoning ordinances are sufficiently binding on housing consumption choices to ensure the validity of the benefit view (Rubinfeld 1987; Ross and Yinger 2000). In addition, although empirical evidence suggests that intrajurisdictional and intrajurisdictional capitalization of differences in property taxes and local expenditures is widespread (Oates 1969; Yinger et al. 1988; and Fischel 2001a, b, who concludes that evidence of ‘capitalization is everywhere’), capitalization is consistent with both the assumption of fixed housing stocks that underlies the benefit tax view and the tax-induced reallocations of capital that underlie the capital tax view (Zodrow 2006); moreover, some observers have argued that in the long run capitalization is inconsistent with the benefit view (Ross and Yinger 2000). Finally, although some recent empirical tests are consistent with the capital tax view (Carroll and Yinger 1994; Wassmer 1993), these results are quite tentative. The primary empirical issue remaining to be resolved is whether the zoning restrictions or other mechanisms stressed by proponents of the benefit tax view are sufficiently binding to preclude the long-run adjustments in housing capital predicted by the capital tax view. This question promises to be a fertile topic for future research, which hopefully will help clarify the answer to the long-standing and critical questions of the incidence and economic effects of the property tax.

See Also