CESifo Economic Studies

CESifo Economic Studies Advance Access originally published online on January 11, 2008
CESifo Economic Studies 2007 53(4):495-521; doi:10.1093/cesifo/ifm019

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 53/4/495    most recent ifm019v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Zodrow, G. R. Search for Related Content Related Collections H10 - General H21 - Efficiency; Optimal Taxation H22 - Incidence H71 - State and Local Taxation, Subsidies, and Revenue Social Bookmarking          What's this?

© The Author 2008. Published by Oxford University Press on behalf of Ifo Institute for Economic Research, Munich. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org

The Property Tax Incidence Debate and the Mix of State and Local Finance of Local Public Expenditures

George R. Zodrow^*

* George R. Zodrow is a Professor of Economics and Rice Scholar, Tax and Expenditure Policy Program, Baker Institute for Public Policy, Rice University, Houston, TX, USA and International Research Fellow, Centre for Business Taxation, Oxford University, e-mail: zodrow{at}rice.edu

	Abstract

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
Many states in the US have in recent years changed the mix of state and local revenue sources used to finance local public expenditures, especially primary and secondary education, with local property taxes being replaced by various sources of state tax revenue. This article examines the desirability of such a tax substitution, focusing on the implications of the long-standing debate between the "benefit tax" and "capital tax" views of the incidence of the tax. It also includes a discussion of some recent research that elaborates the capital tax view of the property tax. (JEL codes: H10, H21, H22, H71)

Key Words: Property tax incidence • capitalization • capital tax view • new view • benefit tax view • Texas tax reform • margin tax

	1 Introduction

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
One of the "new directions in fiscal federalism" that several states in the US, including my home state of Texas, have taken in recent years is a change in the mix of state and local revenue sources used to finance local public expenditures, especially primary and secondary education, with local property taxes being replaced by various sources of state tax revenue. Of course, numerous court cases requiring that the levels of school expenditures be less dependent on local property wealth have played a crucial role in state assumption of a larger share of school finance in many states. Also important have been various property tax limitations, such as those enacted in California and Massachusetts; Texas also has a cap on the property tax rate for revenues used to finance school maintenance and operations.

In addition, more recently, rapid increases in residential housing prices coupled with state education aid rules under which the share of state aid declines as local housing property values increase have resulted in significant increases in the local share of education finance. For example, although house prices in Texas have not experienced growth as rapid as that in many other areas of the country, especially on the east and west coasts, they have increased enough so that the local share of education finance was projected to increase to two-thirds by 2007, up from slightly over one-half in the early-1990s—a local share that would have been the highest among the 50 states (Texas Public Policy Foundation 2003, p. 12; National Education Association 2005, p. 59; Texas Tax Reform Commission 2006, p. 12). In addition, as in many other states, educational expenditures in Texas have been increasing in relative terms. Together, these factors resulted in a significant increase in property tax burdens. For example, residential property tax payments for local school taxes in Texas increased by 118 percent from $4.4 billion in 1997 to $9.6 billion in 2004, significantly outstripping the increase in the level of state GDP growth, which was 47 percent over the same time period (Texas Comptroller of Public Accounts 1999, 2004, 2006).

In Texas, negative public reaction to this increased reliance on the local property tax, coupled with widespread dissatisfaction with the state's "Robin Hood" redistribution plan under which property tax revenues from the school districts that ranked in the top 10 percent in the state, in terms of property per student are redistributed to the remaining districts in the state, was intense (Zodrow 2004). After several years of contentious debate, the result was a sweeping reform in which school property taxes were reduced by a third, replaced by a new state business tax and other sources of revenue that will reduce the local share of primary and secondary education expenses to 50 percent (Texas Tax Reform Commission 2006, pp. 17–20).

A first reaction to these developments might be that they are favorable, as a longstanding tenet of state and local public finance has been that on efficiency grounds it is preferable to assess subnational taxes using jurisdictions that encompass greater areas and larger populations (e.g. state rather than local), as the interjurisdictional mobility of tax bases is significantly lower for such jurisdictions than for smaller entities; as a result, efficiency costs from suboptimal location decisions will be reduced, more progressivity in the tax structure may be introduced if deemed desirable, and economies of scale in revenue collection may be realized. The resulting state revenues can then be redistributed on a formulaic basis, reflecting not only service needs but also both interjurisdictional distributional concerns as necessary and adjustments for positive external benefits; the recipient local governments should then be better equipped to match public services provided to local tastes and should be more accountable to local citizens due to increased transparency and the forces of interjurisdictional competition. As long as the distribution formula is perceived to be equitable (from both a legal and a social perspective) and efficient in the sense that state revenues finance inframarginal expenditures, while local revenues are used for all marginal expenditures (apart from any state subsidies designed to correct for state-level public service externalities at the margin); the conventional wisdom suggests that this tax substitution should be desirable.

This traditional prescription has been challenged in the recent literature. In a recent survey, Oates (2007) notes that excessive reliance on upper-level governments for transfers to finance public local services, including inframarginal units of such services, may lead to the pernicious problem of a "soft budget constraint" under which local governments are periodically "bailed out" of deficit situations; the relevance of this concern is, however, limited in the US to the extent that local governments face binding budget constraints. Oates also notes that empirical work suggests strongly that the design of intergovernmental grant systems in practice falls far short of theoretical ideals, and is instead much more likely to be determined by political considerations. Finally, he observes that intergovernmental grants designed to reduce fiscal inequalities across jurisdictions may impede the migration of capital and employment to lower wage areas that would otherwise narrow these inequalities over time. He concludes that the case against the use of intergovernmental grants rather than own-source revenues at the local level is "not easily dismissed" and argues that the "design and operation of an effective system of intergovernmental grants ranks, in my view, right at the top of the research agenda in fiscal federalism".

Following in the same vein, there is another potential challenge to the traditional argument favoring the use of state-level grants to finance local public expenditures, which is simply that greater reliance on state tax instruments may be undesirable because the typical state revenue sources are relatively less efficient than the property tax finance typically used at the local level. Not surprisingly, a key question in investigating this issue is a resolution of the longstanding debate on the incidence of the property tax, and that debate is the focus of this article, which proceeds as follows.¹ The following section examines, fairly briefly, the advantages and disadvantages of the various commonly used state-level taxes. It concludes that all of these tax instruments suffer from serious problems, so that it is certainly possible that substituting state taxes for local property tax revenues will be undesirable from an efficiency perspective. Section 3 begins by noting that the efficiency properties of the property tax depend primarily on which of two alternative views—referred to as the benefit tax view and the capital tax (or "new") view—more accurately describes the economic incidence of the tax. The remainder of this section, which constitutes the bulk of the article, examines the current state of our knowledge on this crucial tax incidence issue, focusing on recent efforts to reconcile the two views. The concluding section considers directions that future research might take to help narrow the range of opinion on the issue of the incidence of the property tax, and thus help to resolve the directly related issue of whether greater reliance on state-level tax sources coupled with reduced reliance on the local property tax is desirable.

	2 Traditional state revenue options

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
Although, as described earlier, collecting revenues at the state level has certain distinct advantages, one could also argue that an important disadvantage of state-level taxation is that the tax instruments typically used, especially as applied in practice, are not particularly desirable from an efficiency standpoint. Two familiar precepts regarding tax efficiency are especially relevant in the state context.

The first is that the well-known production efficiency theorem argues that taxes on production inputs should generally be avoided.² This in turn suggests that gross receipts taxes are especially undesirable because they result in highly inefficient tax pyramiding, with multiple layers of taxation applied to those products whose inputs are transferred among businesses at various stages in the production process. Under these circumstances, a gross receipts tax imposed at even a very modest rate can become a highly differentiated tax on business inputs that distorts decisions regarding input choices, creates a tax bias toward vertical integration and thus against small businesses including small service providers and, to the extent the tax is shifted forward as higher consumer prices, hampers the export prospects of state businesses and distorts consumer purchasing decisions. However, as will be discussed subsequently, several of the most prominent state tax instruments have the characteristics of a gross receipts tax.

The second general tax policy precept follows from the fact that any state (even Texas!) approximates a small open economy, and thus should treat the after-tax return to capital (and the consumer prices of most tradable goods) as fixed. This of course implies that, apart from taxes levied in accordance with benefits of public services received or to offset negative externalities generated, source-based taxation of mobile capital at the state level is inefficient and indeed counterproductive, as perfectly mobile capital will leave the taxing jurisdiction until the before-tax rate of return to capital invested in the state rises by enough to entirely offset the tax. This in turn implies that local factors of production (land and labor, especially relatively immobile unskilled labor) will bear the full burden of the tax, including its efficiency costs due to capital emigration, a tax bias favoring labor-intensive goods, and any underprovision of public services to local tax competition (Zodrow 2006).

In the following discussion, I first consider how the most commonly used state tax instruments—state sales taxes and state personal and corporate income taxes—fare in terms of these and other efficiency criteria. I then turn to an examination of the intriguing but seldom-used alternative of a state-level value-added tax.

2.1 State sales taxes
In principle, retail sales taxes could be a relatively efficient source of state tax revenue, as they are based on consumption rather than production and do not burden mobile capital. In addition, as a consumption-based tax, sales taxes avoid the distortions of savings and other decisions that characterize the income tax, and could in principle be assessed on a comprehensive and uniform basis.³ In practice, however, existing state retail sales taxes achieve few of these advantages. Because states are concerned about tax evasion in the form of business purchases of consumption goods (and because exemption of all business purchases would be very costly from a revenue standpoint), they are extremely harsh in limiting sales-tax-exempt purchases by businesses. As a result, consumer purchases form a surprisingly small fraction of the typical state sales tax base, estimated by Ring (1999) to be on the order of 60 percent on average. Under such circumstances, the retail sales tax becomes a haphazard form of income tax rather than a consumption tax, and includes elements of both source-based capital income taxation and a gross receipts tax.

Furthermore, the base of the typical state sales tax is far from comprehensive, and is instead characterized by a wide variety of exemptions that are justified on distributional, social, administrative or historical grounds. The situation has been exacerbated by the advent of electronic commerce, since many remote (out-of-state) electronic vendors, like their mail order counterparts, avoid establishing a physical presence or "nexus" in most states and thus avoid any legal responsibility to collect tax on their remote sales. As a result, the efficiency gains from a neutral consumption-based tax are not realized.

Moreover, the most commonly recommended sales tax reform—base broadening by including services that have historically been tax exempt—is not a particularly attractive option. Although base expansion to include untaxed consumer services is desirable from an efficiency perspective, the typical proposal continues to exempt most consumer services and instead involves a great deal of taxation of business services, which would only exacerbate the problems of source-based taxation of capital income and gross receipts taxation described earlier (Hendrix and Zodrow 2003). Thus, replacing local property taxes with revenues raised from increased use of the retail sales tax may not be desirable on efficiency grounds, especially since the sales tax rates in many states are already relatively high, so that further rate increases would significantly exacerbate already large existing distortions.^4,5

2.2 State income taxes
The other main alternative state-level tax is an income tax system consisting of a corporate income tax and a personal income tax. At the individual level, the primary argument against this approach is that it exacerbates all the distortions of the federal income tax. Nevertheless, proponents of the state income tax approach argue that it results in a more equitable tax burden, with straightforward tax exemption of the poor and some progressivity (limited by interstate mobility of high-income individuals), has a relatively comprehensive base, is easy to administer as an add-on to the federal tax, has a base that grows proportionately with the economy, and has the important advantage of providing for residence-based taxation of capital income and is thus not subject to the flaws of source-based capital income taxation noted earlier.

However, at the business level, the corporate income tax is far less attractive. For local corporations, the tax is source-based tax on capital income, replete with distortions of decisions regarding inputs, financing mechanisms, organizational form and retention/payout policy. For corporations with operations in many states, the tax is essentially a tax on the factors in the formula used to apportion the tax base across the states. To the extent that a property component is important, the tax is again a source-based tax on capital income. However, many states now double-weight sales and some, such as Texas, use a single sales factor, which converts the tax to a tax on gross receipts. Indeed, the gross receipts nature of the tax is even more pronounced than with business purchases under the sales tax, since no exceptions are provided (in contrast to the state sales tax, where exceptions are typically provided for material purchased for resale, materials used in the manufacturing or processing of other goods, etc.).⁶

2.3 State value-added taxes
Perhaps the most attractive state-level option is one that is virtually never used in the US—a state-level value-added tax or VAT.⁷ Although there are many alternative structures for such a state VAT, I will focus on the case of an origin-based, consumption-based tax calculated using the subtraction method, sometimes referred to as a "business activities tax" or BAT.⁸ As a consumption-based tax (that allows expensing of all nonlabor business-related purchases, including capital equipment), the BAT largely avoids source-based taxation of mobile capital income; the main exception is that it taxes above-normal returns to capital, which results in an efficient source of revenue in the case of location-specific economic rents, but acts to discourage investment by businesses with highly mobile firm-specific rents. Moreover, because firms are allowed deductions for all business-related purchases, the BAT avoids the problems associated with gross receipts taxes.⁹

The BAT is of course not without its own problems. As a source-based tax on labor income, it creates a disincentive for the hiring of labor, a fatal flaw from the perspective of many politicians, even if taxes on relatively immobile labor are preferable to taxes on highly mobile capital from an efficiency perspective. In addition, businesses are likely to object strenuously to paying taxes when they have no profits—although it should be noted that this argument is inappropriate to the extent taxes paid are viewed as payment for public services received. As a proportional tax, the BAT may be perceived as objectionable on equity grounds, although this concern might be tempered if firms were allowed a minimum deduction per employee, say, roughly equal to the poverty level of income. As a new tax that could not readily be "piggybacked" on an existing federal tax, the BAT would also be relatively expensive from an administrative standpoint (although firms would not have to provide any information not already required for federal income taxes).¹⁰

In summary, none of the common state-level tax instruments, especially as applied in practice, is especially attractive from an efficiency perspective. Moreover, although a state-level VAT such as the BAT may be more desirable on efficiency grounds, experience thus far strongly suggests that it is not a viable political option. A natural question, then, is whether replacing local property taxes with such state-level taxes is likely to be desirable. The answer to this question of course depends on the relative efficiency properties of the property tax, the issue considered next in the analysis.

	3 The Effects of the Property Tax

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
The effects of the property tax, including its economic incidence, efficiency properties and distributional implications, have been the subject of a long and contentious debate, summarized recently in Zodrow (2001a, 2001b), Fischel (2001a, 2001b) and Nechyba (2001). Two alternative views dominate the debate. Under the "benefit tax view", the property tax is an efficient benefit tax paid in exchange for local public services received along the lines envisioned in the celebrated Tiebout (1956) model; it thus has no effects on the distribution of income (Fischel 2001a, 2001b). In marked contrast, under the so-called "new view" of the property tax developed by Mieszkowski (1972)—hereafter, referred to as the "capital tax view"—the property tax, as a tax on the use of capital within a local jurisdiction, causes numerous distortions.¹¹ These distortions include reductions in housing consumption and in the capital intensity of housing, tax-induced misallocations of households and businesses across jurisdictions and reductions in the overall supply of capital to the taxing jurisdiction (Ross and Yinger 1999; Zodrow 2001a, 2001b). Moreover, use of the local property tax can also lead to inefficient underprovision of local public services, as government officials, concerned about tax-induced outflows of mobile capital from their jurisdictions, reduce the level of public services provided (Wilson 1986; Zodrow and Mieszkowski 1986a). The basic tenets of the two views can be summarized as follows.

3.1 The benefit tax view
The benefit tax view argues that the property tax plays the role of the head tax envisioned by Tiebout (1956) in his model of efficiency in the provision of local public services in the presence of perfect consumer mobility and interjurisdictional competition. In the case of the residential property tax, this theory was developed by Hamilton (1975) who followed Tiebout in assuming that individuals sort themselves into local jurisdictions according to demands for public services. He then added the assumptions that these communities utilize binding zoning constraints that establish a minimum house value for each community, and that there are enough communities to accommodate all desired bundles of housing and government services. These assumptions are sufficient to establish an equilibrium in which all communities are homogeneous with respect to both demands for public services and housing; in particular, the zoning constraints preclude individuals from "free-riding" on their neighbors by enjoying local public services while living in a relatively inexpensive home, and the availability of homogeneous high-income communities ensures that no one will choose to live in a relatively expensive home and subsidize local public service provision. In the same tradition, Fischel (1985, 1995) argues that zoning ordinances, defined comprehensively to include a wide variety of land use regulations, are sufficiently restrictive in practice to convert the residential property tax into a benefit tax in the manner envisioned by Hamilton.¹²

Since complete homogeneity of house values within all communities is unrealistic, a natural question is whether the benefit view logic can be extended to communities that are nonhomogeneous with respect to housing values (although still homogeneous with respect to demands for public services). Hamilton (1976) argues that in fact this is the case, if all communities are fully developed (so that housing stocks cannot change in response to the property tax) and homogeneous communities are always available as an option (so that no individual is willing to pay property taxes that exceed benefits received). Under these fairly stringent assumptions, Hamilton shows that capitalization of fiscal differentials into house values converts the property tax into an efficient benefit tax, as relatively expensive (inexpensive) homes sell at a discount (premium) in the heterogeneous community, reflecting the present value of the excess (deficit) of property taxes paid relative to the value of services received.

The implications of the benefit view are striking—the property tax is effectively a user charge that is paid in exchange for the benefits of local public service, and thus distorts neither housing consumption nor the level of public services provided. If the benefit view is accurate, replacing local property taxes with any of the typical state-level tax instruments described above is certain to reduce efficiency. In addition, the benefit view implies that the property tax has no effects whatsoever on the distribution of income.

3.2 The capital tax view
In marked contrast, the "capital tax view" or what was initially called the "new view" of the property tax, developed by Mieszkowski (1972) and extended by Zodrow and Mieszkowski (1986b), argues that the property tax is a differential capital tax that distorts the allocation of capital across local jurisdictions, with its residential component distorting decisions regarding the consumption of housing. Moreover, since the property tax is a source-based tax on mobile capital and local jurisdictions are small open economies, local use of the tax suffers from all the problems of such taxes noted above (Zodrow and Mieszkowski 1983). Furthermore, to the extent that local governments are constrained to use the property tax, they are likely to inefficiently under-provide local public services to avoid driving mobile capital out of their jurisdictions—the familiar tax competition problem described by Oates (1972) and analyzed by Zodrow and Mieszkowski (1986a), Wilson (1986) and Wildasin (1989).¹³

The capital tax view was derived in the context of a general equilibrium model with a fixed national capital stock and two types of local jurisdictions, those characterized by relatively high-tax rates and those characterized by relatively low-tax rates. In this context, Mieszkowski showed that property-tax rates that exceed the national average drive capital out of the high-tax jurisdictions into the relatively low-tax jurisdictions, with opposing effects occurring in relatively low-tax jurisdictions. Property tax differentials thus result in an inefficient misallocation of capital across jurisdictions. In terms of incidence, the "average burden" of all of the property taxes imposed across the nation—the "profits tax effect" of the tax—is borne by capital owners generally. In marked contrast to the benefit tax view, this profits tax effect implies that the property tax is relatively progressive (at least with respect to annual income). In addition, Mieszkowski stressed 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. Since these roughly symmetric excise tax effects tend to cancel, the profits tax effect is the primary factor affecting the distribution of the tax burden under the capital tax view. If the capital tax view is accurate, replacing local property taxes with any of the typical state-level tax instruments will have uncertain effects on efficiency, depending on how the efficiency costs of capital misallocation, housing underconsumption and government service underprovision under the property tax compare to those under the state tax instruments chosen.

Given the dramatically different implications of the benefit and capital tax views, it is perhaps not surprising that there has been considerable skepticism about the possibility of reconciling the two views. For example, Fischel (2001b, p. 170), in arguing for the benefit view, stresses that the "difference between my view and the ‘capital tax’ view promoted by Mieszkowski and Zodrow is profound ... No amount of tinkering with the parameters of the capital-tax view can reconcile it with the approach I have advocated here". Similarly, Mieszkowski (1999, p. xv), in a discussion of the two competing views, concludes that "no reconciliation or hybrid theory as possible". At one level, this is certainly correct. The economic effects of the capital tax view arise due to tax-induced reallocations of housing capital, while such reallocations are precluded under the benefit tax view through either the appropriate zoning constraints or perfect intrajurisdictional capitalization in the context of fully developed communities. Nevertheless, several researchers have constructed hybrid models that include some features of both views. These models can be grouped into three general categories, which are considered in the following three subsections.

3.3 Hybrid models with immobile residents
One problem with the original exposition of the capital tax view by Mieszkowski (1972) is that it was derived within the context of the model of national tax incidence developed by Harberger (1962), and thus ignored many of the aspects of local government use of the property tax stressed by proponents of the benefit view. Zodrow and Mieszkowski (1986a) reformulated the capital tax view to incorporate many of these factors explicitly, including interjurisdictional competition, individual utility functions that include varying tastes for local public services, individuals that are immobile but are segregated into differing communities according to tastes for local public services, and a simple form of land use zoning. Within this context, they show that the national effects of the capital tax view still obtain—as long as capital is mobile across jurisdictions but is fixed in total national supply—and that local jurisdictions tend to inefficiently under-provide local public services due to concerns about tax-induced capital emigration. In addition, following Brown (1924) and Bradford (1978), Zodrow and Mieszkowski (1983) show that the profits tax effect of the capital tax view obtains in the context of a single small taxing local jurisdiction. That is, even though the after-tax rate of return to capital can reasonably be viewed as fixed from the perspective of the taxing jurisdiction, a local tax increase reduces that rate of return very slightly, which, when applied to a large national capital stock, implies that capital bears the full burden of the tax. At the same time, however, the tax-induced outflow of capital from the taxing jurisdiction implies lower returns to relatively immobile factors such as local land and labor and/or higher prices to local consumers. Indeed, for the standard case in which the local jurisdiction is a small open economy, the tax burden borne by local factors, in the form of lower wages and land rents or higher prices, approximates the total burden of the tax (Kotlikoff and Summers 1987).

The implication of the latter point is that even though these models clearly obtain capital tax view results, they also have a striking benefit tax view flavor. Specifically, even though the property tax causes inefficiencies in the form of capital misallocation and government service underprovision, the burden of increases in local government expenditures financed with increases in the local property tax is borne entirely by local residents, an essential characteristic of the benefit tax view.

Although these models establish that the burden of the property tax is borne by local residents, they do not derive the distribution of that burden between labor and land (since both are fixed), between factor owners and consumers and among consumers of different goods. In ongoing research, Athiphat Muthitacharoen and I are investigating this issue in the context of a two-sector (e.g. housing and a tradable good) open-economy model in which labor is mobile across sectors (although fixed in total supply within the jurisdiction) while land is immobile.¹⁴

Even in this relatively simple model, the incidence of the property tax is fairly complex, primarily because labor is mobile across the two production sectors and because forward shifting can occur only in the nontradable good sector. As expected, land in the two sectors bears the full burden of the land component of the property tax, and a significant fraction of the capital portion. However, all of the other effects of the tax are theoretically ambiguous. In particular, the effect of the tax on wages is ambiguous, as the property tax drives out mobile capital, which tends to reduce labor productivity and thus wages but, since is labor is mobile across sectors, the tax is more likely to be borne by land owners in the two production sectors and by consumers in the nontradable goods sector.

Given these theoretical ambiguities, we simulate the effects of introducing a property tax in the model. Preliminary results suggest that the property tax falls primarily on landowners in the two production sectors. In addition, some forward shifting to consumers of the nontradable good occurs, on the order of 10–40 percent of the burden of the tax. But the most striking result is that typically very little or none of the burden of the property tax is borne by labor; indeed, wages typically increase when the property tax increases, primarily due to over-shifting of the tax to land as well as some increases in the price of the nontradable good. This result obtains despite the fact that there is capital flight from both production sectors in response to the imposition of the business property tax.¹⁵

3.4 Hybrid models with perfectly mobile residents
Although the models in the previous section have many of the features that characterize the benefit tax view approach, including households that are segregated according to their demands for public services, they do not allow for the perfectly mobile households assumed by both Hamilton and Tiebout. Perfect consumer mobility is a central feature of the models of the property tax constructed by Hoyt (1991), Krelove (1993) and Wilson (1997), which, in the case of identical individuals, implies that all households must attain the same level of utility. As described in an excellent recent paper synthesizing this work by Wilson (2003), this line of research stresses that with perfect consumer mobility, the property tax has an important benefit tax aspect in that the property tax payment by each household within a jurisdiction equals the marginal cost of providing it with local public services. Nevertheless, in these hybrid models, the property tax is, as Wilson (2003, p. 220) describes it, a "distortionary benefit tax" in that consumption of both housing and local public services are set at inefficiently low levels.

More specifically, when head tax finance is available in these models, land value maximizing governments ensure an efficient level of public services by setting the head tax equal to marginal congestion costs. [In addition, if there are economies of scale in local public service provision so that average costs exceed marginal costs, a land tax should be used to balance the budget as described in Flatters, Henderson and Mieszkowski (1974); I shall, however, ignore this complication on the grounds that most empirical evidence suggests that the services provided by local governments are largely constant cost "publicly provided private goods".¹⁶] Surprisingly, when head tax finance is replaced with property tax finance, the marginal tax payment still equals the marginal cost of providing a new household with public services, as occurs under the benefit tax view—that is, the property tax serves as an indirect form of congestion fee. Nevertheless, the property tax still distorts housing consumption decisions (there is no zoning in these models) and results in underprovision of local public services, and the capital tax portion of the property tax is borne by capital—all features that are clearly characteristic of the capital tax view of the property tax.¹⁷

3.5 A model of intrajurisdictional capitalization
The hybrid models considered thus far are all characterized by local jurisdictions that are comprised of identical households. However, as described earlier, the more realistic of the two models constructed by Hamilton provides for jurisdictions that are heterogeneous with respect to housing while still homogeneous with respect to demands for public services, with perfect capitalization of fiscal differentials into the prices of a fixed nonhomogeneous housing stock converting the property tax into a benefit tax. Indeed, in making his case that the combination of comprehensive zoning and fiscal capitalization converts the property tax into a benefit tax, Fischel (2001a, 2001b) argues that that the extensive literature on the capitalization of property taxes and public services implies that fiscal "capitalization is everywhere" (Fischel 2001a, p. 56). He goes on to conclude that empirical support of fiscal capitalization provides compelling evidence that the benefit tax view accurately describes the effects of the property tax.

In some ongoing research (Zodrow 2007), I am examining the contention that the phenomenon of fiscal capitalization in heterogeneous communities confirms the validity of the benefit tax view. Of course, the discussion in Section 3.3 makes it clear that interjurisdictional capitalization is consistent with both views of the property tax. That is, the small open economy interpretation of the capital tax view implies that the burden of a local property tax will be borne by local factors of production in the jurisdiction, especially land if labor is at least partially mobile—that is, property taxes across jurisdictions will be capitalized into land values. The goal of this research project is to determine whether, as suggested without proof by Mieszkowski and Zodrow (1989), the same result obtains for the intrajurisdictional capitalization of fiscal differentials stressed by Hamilton (1976).

Like the other work described earlier, the structure of the model in Zodrow (2007) is chosen to be consistent with the assumptions of the Tiebout–Hamilton tradition. In particular, a limited form of land use zoning (fixed amounts of land for high and low value housing, the two types of housing in the model) is introduced so that land value capitalization is a possibility. This assumption is essential since, as stressed by Ross and Yinger (1999), in the absence of zoning all land within a jurisdiction would in the long run sell for the same price so that capitalization would be impossible.¹⁸ In addition, all households are perfectly mobile with an exogenous source of income and an exogenously specified level of welfare, there are two types of households differentiated by their demand for housing, and there are enough jurisdictions to satisfy all tastes for local public services. In the initial equilibrium, head taxes are allowed so that efficiency in both housing consumption and local public service provision is achieved. Property taxes on all land and capital within the jurisdiction are then introduced, with the revenues used to reduce the level of head taxation, holding the level of public services per capita fixed.¹⁹

The main result of this differential tax incidence analysis is that, evaluated at the initial head tax equilibrium (so the excess burden effects are zero), the land price capitalization effects under the capital tax view are identical to those under the benefit tax view, despite the very different mechanisms giving rise to these capitalization effects under the two views. Under the benefit view, these capitalization effects simply reflect the differentially high (low) property taxes paid by high (low) value homes, holding housing capital stocks fixed. By comparison, under the capital tax view, capitalization effects arise because households alter their housing consumption in response to the imposition of the property tax. Specifically, capital is reallocated out of the production of large houses, where property taxes are high relative to benefits received and into production of smaller homes, where the property tax bill is low relative to benefits received. As a result, land rents unambiguously increase for land used for small houses, and unambiguously decrease for land used for large houses, and it is these changes in land rents that are capitalized into land prices.

Since these results imply that land value capitalization effects can be precisely the same under the two views, they suggest that empirical evidence of intrajurisdictional capitalization does not allow one to distinguish between the validity of the benefit tax and capital tax views of the property tax.²⁰ Nevertheless, as in the other hybrid views of the property tax described earlier, local use of the property tax has some of the characteristics of a benefit tax, as residents pay for net local public services received (those not financed with head taxes) in the form of higher housing prices. Moreover, since fiscal differentials are capitalized into land values, the net effect of the property tax burden and land value capitalization of the fiscal differential is that in equilibrium both types of households effectively pay only for services received.²¹ Thus, as in the hybrid models described above, residents "pay for what they get" under both views. However, under the capital tax view, land value capitalization occurs due to capital reallocations across housing types, implying inefficiency in the housing market. By comparison, under the benefit tax view, capitalization occurs with respect to fixed housing capital stocks, and there is no distortion of the allocation of housing capital.

	4 Conclusion

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
Although the pure versions of the benefit and capital tax views of the property tax are fundamentally incompatible, the various hybrid views discussed in this article—all of which, in my view, are considerably closer to the capital tax view than to the benefit tax view—represent important steps toward reconciling the long-standing debate in the literature about which view more accurately describes the effects of the property tax. All of the hybrid views, however, deviate from the benefit tax view in that they assume that zoning does not impose binding constraints on the consumption of housing at the margin, so that the property tax distorts housing consumption decisions.

At least two avenues for future empirical research are promising. The first is a determination of the extent to which actual zoning constraints are binding on marginal housing decisions, so that the property tax cannot inefficiently distort housing decisions. Fischel (1992, 1995) stresses that the practice of zoning is widespread, although Glaeser and Gyourko (2002) argue that zoning has only minimal impacts on housing supply in many regions of the country, as its effects are concentrated in California and several eastern cities.²² Fischel also notes that a wide variety of zoning restrictions are utilized—including minimum lot sizes (and one home per lot), setbacks, height restrictions, requirements for off-street parking, restrictions to single family use, minimum square footage, differential building code requirements, the designation of certain areas as off limits for environmental or other reasons and requirements for the provision of infrastructure at the expense of the developer—and argues that they are sufficient to ensure the validity of the benefit tax view.

Nevertheless, as stressed by Mieszkowski and Zodrow (1989) and Ross and Yinger (1999), the Hamilton (1975) version of the benefit view obtains only if zoning requirements are binding at the margin—that is, "perfect zoning" is required for the benefit view to be fully operative. Of course, as noted by Ladd (1998, p. 34), perfect zoning in all cases would never be expected; rather the issue is whether zoning under the benefit view "sufficiently approximates reality that it becomes useful for making predictions and drawing conclusions" about the incidence of the property tax—in particular whether zoning reduces the efficiency losses that might otherwise occur under the property tax by limiting its distortions of housing consumption. Unfortunately, evidence on the extent to which zoning constraints are binding—and thus limit the potential distortions of the property tax—is extremely difficult to obtain. Ladd notes that "no one would disagree that the property tax would distort decisions about minor expansions and repair that are beyond the purview of the zoning authority but not the tax assessor"—that is, one would not expect the benefit view to be operative at the margin for such changes in the housing capital stock. Similarly, even the most ardent proponents of the benefit view do not assert that it is operative in large and highly heterogeneous urban areas.²³ But the extent to which zoning constraints are binding in the suburban communities that are the focus of the Tiebout–Hamilton analysis is difficult to determine empirically. Indeed, in contrast to the position taken by Fischel, various observers have argued that zoning ordinances, no matter how prevalent, are not likely to be sufficiently binding for the benefit view to obtain. For example, Ross and Yinger (1999) note that many studies have demonstrated considerable heterogeneity with respect to house values and income levels in suburban jurisdictions, suggesting that zoning constraints are typically not binding, and argue that "zoning tools, such as lot size restrictions and set-back rules, appear far too blunt to control H (housing) precisely".²⁴ Similarly, Rubinfeld (1987, p. 591) concludes that "there is reason to believe that actual zoning policies deviate substantially from the one which transforms a property tax into a head tax". Nevertheless, these conjectures do not take the place of the admittedly onerous task of conducting a detailed property-by-property study to determine the extent to which the zoning requirements impose binding constraints on marginal housing consumption decisions.

A second approach would be to investigate the extent to which relatively high property taxes reduce housing consumption in practice or reduce the capital intensity of housing production (and nonhousing production as well), as predicted by the capital tax view.²⁵ Wassmer (1993) provides some intriguing evidence on these questions. He constructs a sample of 62 cities and tests whether positive property tax differentials, relative to the average level of taxation for the sample, reduce the capital intensity of production or are reflected in lower property values in due to capitalization effects. Although his results are generally supportive of the capital tax view, the effects are quite modest. Moreover, Wassmer's sample is cities that are typically geographically separated, rather than the suburban communities that are the focus of the Tiebout–Hamilton benefit tax view.

More recently, an empirical study by Lutz (2006) examines the effects of a school finance reform in New Hampshire that significantly reduced property taxes. He shows that in general there was a significant increase in housing investment in the state over the 5 years after the property tax reduction, consistent with the capital tax view, with an implied investment elasticity of roughly one. However, at the same time, Lutz also finds no evidence of an investment response in the suburban ring surrounding Boston, where the property tax reduction was instead largely capitalized into house values, consistent with the benefit tax view, a result that he attributes to a combination of limited land for new development and zoning. These results reinforce the idea that the fixity of housing is the critical issue in thinking about the incidence of the property tax (Nechyba 2001; Zodrow 2007). In this context, it should be noted that the relative inelasticity of the supply of housing, while important, should not be overstated. The required responses in capital allocation can occur over time through a combination of changes in new construction, renovations including teardowns, and depreciation and new investment. Moreover, much of the capital intensity of the existing housing capital stock may already reflect property tax induced underconsumption, as past housing decisions were based on expectations about the extent of property tax finance of local public services. In any case, further empirical investigation of the effects of the property tax on housing capital intensity and housing investment would appear to be fruitful areas for future research.²⁶

Turning next to fiscal federalism issues, the relative desirability of replacing local property taxes with state-level taxes obviously depends on the state-level instrument chosen. For example, in the Texas case, despite a long debate regarding numerous reasonable potential tax instruments, the final outcome was a so-called "margin tax", which can loosely be described as a modified gross receipts tax, where firms get a choice between (i) a deduction for the cost of materials including depreciation, (ii) a deduction for the cost of labor and (iii) a deduction equal to 30 percent of gross receipts, with no additional deductions. The tax is thus a haphazard form of gross receipts tax that shares most of the problems of such taxes (including effective tax rates that will in many cases significantly exceed the relatively low nominal rates of tax), and a variety of other problems as well. The new Texas margin tax is thus likely to be highly distortionary, not to mention the obvious incentives it creates for organizing business activities so that one entity is labor-intensive and chooses to be taxed under the component of the margin tax that allows deductions for labor compensation, while another entity takes maximum advantage of the deduction for cost of goods sold.²⁷ If such a state tax instrument is used to replace local property tax revenues, the case for such a reform becomes exceedingly weak.

More generally, the desirability of replacing local property taxes with state tax revenues should be analyzed in the context of a dynamic general equilibrium model that can weigh the relative efficiency costs of the various tax substitutions contemplated. Such a model would have to have a rich business structure, so that it could capture the production inefficiencies associated with sales taxes that impinge significantly on business input purchases and with gross receipts taxes that create tax burdens that vary with the degree of vertical integration. Again, such investigations would appear to be a promising avenue for future research. Note, however, that replacing local property taxes with any of the typical state tax instruments has one serious disadvantage. Specifically, a primary distortion of the property tax is that tends to reduce housing consumption. However, this feature of the tax may simply offset the tax bias favoring the consumption of housing due to the fact that the federal income tax does not tax the imputed rent on owner-occupied housing, while allowing deductions for home mortgage interest (and for property taxes). Thus, it is possible that the housing consumption "distortion" associated with the property tax is actually efficiency enhancing. On the other hand, state use of corporate and individual income taxes tends to exacerbate the many distortions of the federal income tax system, and state sales taxes are likely to have similar effects, especially with respect to decisions regarding labor supply. Thus, on efficiency grounds, there is a presumption that replacing local property taxes with state-level taxes is likely to reduce economic efficiency, a presumption that can be reversed only with explicit modeling of the efficiency gains obtained in other areas from such a tax substitution.

	Acknowledgements

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 
I would like to thank Eugenia Toma, Peter Mieszkowski, the participants at a conference on "New Directions in Fiscal Federalism", sponsored by the Institute for Federalism and Intergovernmental Relations, University of Kentucky and CESifo, held in Lexington, Kentucky, 14–16 September 2006, and two anonymous referees for helpful comments and Athiphat Muthitacharoen and Joyce Tung for excellent research assistance. Much of the discussion in this article draws on research supported by the Lincoln Institute of Land Policy.

	Footnotes

 
¹ Of course, the same revenue sources could in principle be used at either the state or local level; for example, a statewide tax on nonresidential property may be an attractive option relative to local use of the tax, and some local governments utilize income taxes. Nevertheless, the following analysis assumes that, for historical, political or other reasons, local governments rely on the local property tax and state governments rely on other taxes, primarily individual and corporate income taxes, sales taxes, asset taxes and other commonly used state tax instruments.

² The Diamond and Mirrlees (1971) production efficiency theorem states that—under the appropriate circumstances—taxes on business inputs should be avoided entirely. The basic intuition behind this result is that the appropriate set of taxes on consumption goods alone can achieve any outcome that would obtain under taxation of production inputs, but consumption taxes avoid the distortions of input choices that arise with production taxes. This result is subject to a number of qualifications; in particular, it requires that commodity taxes be set optimally and that any above-normal profits be subject to tax (Slemrod 1990) and the theorem, derived in a closed economy context, applies only under certain conditions in a multi-jurisdictional context (Keen and Wildasin 2004). Nevertheless, the production efficiency theorem suggests that, at least to a first approximation, taxes on business inputs should be avoided, especially if—as seems inevitable—they are not explicitly designed to offset any problems that arise under the system of taxation of final goods in the state.

³ Although uniform taxation is not necessarily efficient, it is likely to represent the most practically feasible approximation to an efficient commodity tax structure. In addition, sales taxes are destination based, and are thus at least loosely consistent with the benefit principle, which suggests that state residents, who receive the benefits of state public services, should finance their provision.

⁴ Note that another argument against such a reform is that the property tax is fully deductible against federal personal income tax liability while state sales taxes may not be deductible.

⁵ A more promising, but far less politically feasible, approach would be fundamental reform of the sales tax, involving expansion of the base to include as many consumption items as possible while offsetting the distributional consequences of this approach with a means-tested rebate, coupled with full exemption of all business purchases from tax (McLure 2000).

⁶ Some states also utilize various forms of asset taxes, which are also subject to the criticism that they are a form of source-based capital income taxation and, if utilized as a minimum tax in combination with a corporate income tax, create a tax bias against risk-taking, since some firms will be subject to the income tax if a risky venture is successful, but will be subject to the asset tax if their venture fails and they fall into a loss situation. Statewide taxation of property is also an option, but one that is seldom used.

⁷ In the US, only New Hampshire and Michigan utilize forms of a state-level VAT, and the Michigan tax is rapidly being phased out.

⁸ Note that such a tax would in general share the efficiency properties of any consumption-based tax; for recent discussions of the still controversial issue of the relative efficiency properties of national consumption and income taxes, see the articles in Zodrow and Mieszkowski (2002), Aaron, Burman and Steuerle (2007) and Diamond and Zodrow (2007).

⁹ In addition, the BAT can be justified as a proxy for a benefit related tax on businesses, assuming that this is demand for public services is roughly proportional to the value added attributable to production in the state (Cline 1988; Bird 2000), although this rationale suggests that an income-based state VAT would be the appropriate tax instrument.

¹⁰ Note also that the BAT would have to be apportioned for multistate firms. The most logically consistent approach would be to use payroll or total labor compensation as the apportionment factor, since the BAT is primarily a tax on labor. Alternatively, both payroll and property could be used in the apportionment formula on the grounds that the BAT taxes above normal returns to capital, as well as returns to existing capital (Zodrow 2004).

¹¹ A third "traditional view" of the property tax, which argues that the property tax is fully shifted forward to consumers in the form of higher prices, can be shown to be a special case of the capital tax view (Wildasin 1986).

¹² In addition, Fischel (1975) and White (1975) argue that the nonresidential property tax can also be viewed as a benefit tax in the presence of interjurisdictional competition for firms, the appropriate zoning restrictions, and one-on-one negotiations between firms and government officials.

¹³ Note, however, that interjurisdictional tax competition may also generate efficiency benefits, especially in restraining tendencies toward over-consumption of government services, as described in Wilson (1999), Zodrow (2003) and Wildasin and Wilson (2004).

¹⁴ See Muthitacharoen and Zodrow (2006, 2007). This work builds on the analysis of Sullivan (1984) who analyzes a similar model but assumes that labor is perfectly mobile across metropolitan areas.

¹⁵ This result is in the same vein as those obtained by Gravelle and Smetters (2006) in their analysis of the incidence of the US corporate tax in a world economy, although they focus on the roles of imperfect substitution in product markets and imperfectly mobile capital.

¹⁶ In addition, further complications—also ignored in this analysis—arise if individuals are sufficiently heterogeneous that economies of scale cannot be exhausted within homogeneous communities and individuals must be partitioned across communities (Henderson 1991).

¹⁷ However, Wilson (2003) shows that the congestion tax feature of the property tax implies that the degree of underprovision of local public services is less under the property tax than it would be under a land tax.

¹⁸ For example, such an equilibrium, with all land prices equalized, characterizes the model constructed by Wilson (2003).

¹⁹ An obvious direction for future research is to make the level of public services endogenous, as in Zodrow and Mieszkowski (1986a, 1986b) and Wilson (2003).

²⁰ In addition, housing consumption declines unambiguously for both types of households, requiring that total population in the jurisdiction must increase and, as in Hamilton (1976), the net effect of all property tax capitalization effects on aggregate land value within the taxing jurisdiction is zero.

²¹ It is important to note, however, that this result obtains only in equilibrium, that is, for subsequent purchasers of the housing after the effects of a tax increase are capitalized into land values. By comparison, when the tax is initially imposed, the land value capitalization effects that occur at the time of implementation do not reflect benefits received and may cause a wide variety of distortions—a result that is clearly a serious weakness of the benefit tax view interpretation of the implications of perfect capitalization of fiscal differentials. For discussions of the efficiency implications of these capitalization effects in the context of school finance equalization measures, including effects on expenditure levels, see Hoxby (2001) and Hoxby and Kuzeimko (2004).

²² This suggests a separate test—if the benefit view holds, then areas with relatively strict zoning should be more likely to use the property tax, since it more closely approximates a benefit tax. Since some support for this proposition is provided by the fact that the northeastern states in general obtain a relatively large fraction of total state and local revenues from the property tax (Zodrow 2004), a systematic investigation of this argument would also be a useful area for further research. Note that although California, as a relatively high-zoning but low property tax state, is an exception to this rule, Glaeser (1996) attributes this to court-mandated uniformity in educational expenditures, the resulting property tax limitations and rapidly increasing land prices, all of which broke the link between property taxes and land values. More generally, analyses of trends in property tax use including property tax limitations, especially in the light of court mandates to reduce reliance on the property tax as a means of financing public education, may provide some indirect evidence on the perceived desirability of the property tax as a revenue instrument, relative to alternative taxes.

²³ For example, Ladd (1998, pp. 34–35) notes that the benefit view is also not likely to obtain in exurban or rural areas, so that under any circumstances for a "significant proportion of the US population, the property tax is not appropriately viewed as a benefit tax". Fischel (1995) agrees with this assessment and describes empirical evidence that demonstrates a variety of differences across suburban and central city communities, which suggests that the assumptions underlying the benefit view are satisfied only in the suburbs.

²⁴ Ross and Yinger (1999) also stress the zoning is consistent with alternative models of housing and local public goods determination.

²⁵ Interpretations of the empirical evidence on the debate very considerably. For example, Ross and Yinger (1999) argue that "the evidence against the benefits view is overwhelming", while Oates (2001, p. 142) concludes that "As things stand, it is impossible to reject either the new view or the benefits view in favor of the other".

²⁶ Housing rents provide a limited test of the two views—forward shifting of the property tax in housing rents would be consistent with both views, but backward shifting to property owners would be consistent with the capital tax view. Although numerous early studies yielded inconclusive results on this issue, Carroll and Yinger (1994) argue that these studies suffer from numerous methodological problems. They analyze a sample of 147 towns and cities in the Boston metropolitan area, correct for the methodological problems noted above, and find that landlords bear a large fraction of property tax increases in these communities, consistent with the capital tax view of the property tax.

²⁷ Consolidation rules under the Texas margin tax were recently strengthened to limit such avoidance schemes.

	References

Top Abstract 1 Introduction 2 Traditional state revenue... 3 The Effects of... 4 Conclusion Acknowledgements References

 

Aaron HJ, Burman L, Steuerle CE. Taxing Capital Income. (2007) Washington, DC: Urban Institute Press.

Bird RM. "Rethinking Subnational Taxes: A New Look at Tax Assignment". Tax Notes International (2000) 8:2069–96.

Bradford DF. "Factor Prices May Be Constant but Factor Returns are not". Economics Letters (1978) 1:199–203.[CrossRef][ISI]

Brown HG. The Economics of Taxation. (1924) New York: Holt.

Carroll R, Yinger J. "Is the Property Tax a Benefit Tax? The Case of Rental Housing". National Tax Journal (1994) 47:295–316.

Cline RJ. "Should States Adopt a Value-Added Tax?" In: The Unfinished Agenda for State Tax Reform.—Gold SD, ed. (1988) Denver, CO: National Conference of State Legislatures.

Diamond JD, Zodrow GR. Fundamental Tax Reform: Issues, Choices and Implications. (2007) Cambridge, MA: MIT Press. (forthcoming).

Diamond PA, Mirrlees J. "Optimal Taxation and Public Production". American Economic Review (1971) 61:8–27. and 261–78.[ISI]

Fischel WA. "Fiscal and Environmental Considerations in the Location of Firms in Suburban Communities". In: Fiscal Zoning and Land Use Controls.—Mills ES, Oates WE, eds. (1975) Lexington, MA: Lexington Books. 119–73.

Fischel WA. The Economics of Zoning Laws: A Property Rights Approach to American Land Use Controls. (1985) Baltimore: Johns Hopkins University Press.

Fischel WA. "Property Taxation and the Tiebout Model: Evidence for the Benefit View from Voting and Zoning". Journal of Economic Literature (1992) 30:171–7.[ISI]

Fischel WA. Regulatory Takings: Law, Economics, and Politics. (1995) Cambridge, MA: Harvard University Press.

Fischel WA. "Municipal Corporations, Homeowners and the Benefit View of the Property Tax". In: Property Taxation and Local Government Finance.—Oates WE, ed. (2001) Cambridge, MA: Lincoln Institute of Land Policy. 33–77.

Fischel WA. "Homevoters, Municipal Corporate Governance, and the Benefit View of the Property Tax". National Tax Journal (2001b) 54:157–73.

Flatters F, Henderson V, Mieszkowski P. "Public Goods, Efficiency and Regional Fiscal Equalization". Journal of Public Economics (1974) 3:99–112.[CrossRef]

Glaeser E. "The Incentive Effects of Property Taxes on Local Governments". Public Choice (1996) 89:93–111.[CrossRef][ISI]

Glaeser E, Gyourko J. "The Impact of Zoning on Housing Affordability". (2002) Harvard Institute of Economic Research Discussion Paper 1948.

Gravelle JG, Smetters KA. "Does the Open Economy Assumption Really Mean That Labor Bears the Burden of a Capital Income Tax". Advances in Economic Policy (2006) 6:1–42.

Hamilton BW. "Zoning and Property Taxation in a System of Local Governments". Urban Studies (1975) 12:205–11.[Free Full Text]

Hamilton BW. "Capitalization of Intrajurisdictional Differences in Local Tax Prices". American Economic Review (1976) 66:743–53.[ISI]

Harberger AC. "The Incidence of the Corporate Income Tax". Journal of Political Economy (1962) 70:215–40.[Medline]

Henderson JV. "Separating Tiebout Equilibrium". Journal of Urban Economics (1991) 29:128–52.[CrossRef][ISI]

Hendrix ME, Zodrow GR. "Sales Taxation of Services: An Economic Analysis". Florida State Law Review (2003) 30:411–33.

Hoyt WH. "Competitive Jurisdictions, Congestion, and the Henry George Theorem: When Should Property be Taxed Instead of Land?" Regional Science and Urban Economics (1991) 21:351–70.[CrossRef][ISI]

Hoxby CM. "Not All School Finance Equalizations are Created Equal". Quarterly Journal of Economics (2001) 116:1189–231.[CrossRef][ISI]

Hoxby CM, Kuzeimko I. "Robin Hood and His Not-So-Merry Plan: Capitalization and the Self-Destruction of Texas’School Finance Equalization Plan". (2004) Cambridge, MA: National Bureau of Economic Research Working Paper W10722, National Bureau of Economic Research.

Keen M, Wildasin D. "Pareto-Efficient International Taxation". American Economic Review (2004) 94:259–75.

Kotlikoff LJ, Summers LH. "Tax Incidence". In: Handbook of Public Economics.—Auerbach AJ, Feldstein MS, eds. (1987) North Holland: Amsterdam. 1043–92.

Krelove R. "The Persistence and Inefficiency of Property Tax Finance of Local Public Expenditures". Journal of Public Economics (1993) 51:415–35.[CrossRef][ISI]

Ladd HF. Local Government Tax and Land Use Politics in the United States. (1998) Northhampton, MA: Edward Elgar Publishing.

Lutz BF. "How Elastic is Residential Capital Investment to Property Taxation? Evidence from Court Induced Tax Changes". (2006) Paper presented at the 99th Annual Conference on Taxation, National Tax Association, November 16–18, 2006: Boston MA.

McLure CE Jr. "Rethinking State and Local Reliance on the Retail Sales Tax: Should We Fix the State Sales Tax or Discard It?" Brigham Young University Law Review (2000) :77–137.

Mieszkowski P. "The Property Tax: An Excise Tax or a Profits Tax?" Journal of Public Economics (1972) 1:73–96.[CrossRef]

Mieszkowski P, ed. "Introduction". In: Taxes, Public Goods and Urban Economics: The Selected Essays of Peter Mieszkowski. (1999) Cheltenham, UK: Edward Elgar.

Mieszkowski P, Zodrow GR. "Taxation and the Tiebout Model: The Differential Effects of Head Taxes, Taxes on Land Rents, and Property Taxes". Journal of Economic Literature (1989) 27:1098–146.[ISI]

Muthitacharoen A, Zodrow GR. "State and Local Taxation of Business Property: A Small Open Economy Perspective". In: Proceedings of the Ninety-Eighth Annual Conference on Taxation. (2006) 434–40. National Tax Association, 2006.

Muthitacharoen A, Zodrow GR. "The Excise Tax Effects of a General Property Tax". (2007) Rice University. unpublished manuscript.

National Educational Association. Rankings and Estimates: Rankings of the States 2004 and Estimates of School Statistics 2005. (2005) (last accessed 21 December 2007). http://www.nea.org/edstats/images/05rankings.pdf.

Nechyba TJ. "The Benefit View and the New View: Where Do We Stand, Twenty-Five Years into the Debate?" In: Property Taxation and Local Government Finance.—Oates WE, ed. (2001) Cambridge, MA: Lincoln Institute of Land Policy. 113–21.

Oates WE. Fiscal Federalism. (1972) New York: Harcourt Brace Jovanovich.

Oates WE. "Property Taxation and Local Government Finance: An Overview and Some Reflections". In: Property Taxation and Local Government Finance.—Oates WE, ed. (2001) Cambridge, MA: Lincoln Institute of Land Policy. 21–31.

Oates WE. "On the Theory and Practice of Fiscal Decentralization". In: Key Issues in Public Finance.—Auerbach Alan J, Shaviro Daniel, eds. (2007) Harvard University Press. (forthcoming).

Ring RJ Jr. "Consumer's Share and Producer's Share of the General Sales Tax". National Tax Journal (1999) 52:79–90.

Ross S, Yinger J. "Sorting and Voting: A Review of the Literature on Urban Public Finance". In: Handbook of Regional and Urban Economics.—Cheshire P, Mills ES, eds. (1999) Vol. 3, North Holland: Amsterdam, 2001–60.

Rubinfeld DL. "The Economics of the Local Public Sector". In: Handbook of Public Economics.—Auerbach AJ, Feldstein MS, eds. (1987) North Holland: Amsterdam. 571–645.

Slemrod J. "Optimal Taxation and Optimal Tax Systems". Journal of Economic Perspectives (1990) 4:157–78.

Sullivan A. "The General Equilibrium Effects of the Industrial Property Tax: Incidence and Excess Burden". Regional Science and Urban Economics (1984) 14:547–63.[CrossRef][ISI]

Texas Comptroller of Public Accounts. Statements: 1998 School Property Taxes: Who Paid and How Much? (1999) (last accessed 21 December 2007). http://www.window.state.tx.us/taxinfo/proptax/stmt/stmt9904/stmt9904a.html.

Texas Comptroller of Public Accounts. Annual Property Tax Report: Tax Year 2004; Part 2: Focus on School Taxes. (2004) (last accessed 21 December 2007). http://www.window.state.tx.us/taxinfo/proptax/annual04/focus.html.

Texas Comptroller of Public Accounts. Spring 2006 State Economic Forecast: Real Gross State Product. (2006) (last accessed 21 December 2007). http://www.window.state.tx.us/ecodata/fcst06spr/.

Texas Public Policy Foundation. Putting the Sides Together—Twelve Perspectives on Texas Public School Finance. (2003) (last accessed 21 December 2007). http://www.texaspolicy.com/PTST/.

Tax Fairness: Property Tax Relief for Texans. (2006) (last accessed 21 December 2007). http://www.governor.state.tx.us/priorities/tax_reform/TTRC_report/files/TTRC_report.pdf.

Tiebout CM. "A Pure Theory of Local Expenditures". Journal of Political Economy (1956) 64:416–24.