Beyond Sales Price: Wealth Measures and Property Tax Inequities
Beyond Sales Price: Wealth Measures and Property Tax Burdens
Prepared For
Submission to the
Journal of Real Finance and Economics
Submitted By
Brent C Smith
Assistant Professor
Department of Finance, Insurance and RealEstate
VirginiaCommonwealthUniversity
School of Business
P.O. Box 844000
Richmond, VA23284
804 828 7161
Abstract
The results from this study extend the academic literature on vertical inequities in property taxation by analyzing intrametropolitan spatial patterns in the property tax burden on residents of Chicago, Illinois. Property transaction data from the Cook County Assessor’s Office is observed over an eleven-year period. An expansion model is created to test for spatial vertical inequity in the property tax burden. In a second step, individual property tax burdens regressed on structural and locational variables with the goal of providing expanded detail on tax inequities beyond the value of the property. The findings are vitally important to policymakers interested in funding urban redevelopment programs targeting wealth building through real estate ownership in low-income neighborhoods, and equally important to analyst concerned with market segmentation and the implications of tax capitalization on real estate markets.
Key Words: residential real estate, property taxation, public economics
Introduction
Though the property tax has been maligned throughout much of its history (see Lutz, 1918), it has been, and remains, the major source of tax revenue for local governments in the United States. As home prices and property taxes in many areas of the U.S. continue to reach new heights, homeowners are aiming their sights at a common target: the local tax assessor. Couple this with the fact that households are increasingly consuming more of their disposable income on housing; the apparent increase in the property tax burden is stimulating a new revolt by homeowners. The scapegoat in all this is often the assessor, and across the country assessors are facing suits and calls for their firing from local citizens.
The assessment quality is affected, in part, by differences in market environments within property tax jurisdictions, and such differences that are interpreted by citizens as assessor incompetence may actually be variable market forces.The result of this heightened interest has been the creation of property tax restructuring legislation in California, Michigan, Massachusetts, Florida, Indiana and under consideration in Illinois where rapid expansion of tax increment financing is limiting the fiscal control of local jurisdictions to provide public services (Birch, Sunderman, Smith, 2004).
In the property tax literature, an ad valorem property tax is considered equitable if all properties in the taxing jurisdiction are subject to the same effective tax rate. That is, all properties, regardless of value or type, should be taxed at the same percentage of their market value. Because market value is a theoretical construct and not directly observable, errors in estimating market value may result in systematic inequity, with some properties taxed at higher effective rates than others (Allen and Dare, 2002). A number of models have been developed to test for the presence of inequities in the assessment process (Cornia and Slade, 2005; Smith, Sunderman, Birch., 2003; Smith, 2002, Smith, 2000). Further, there has been an effort to illustrate that tax inequities can exhibit spatial variations (Smith, 2002; Thrall, 1998). To date, the research on property tax inequities focuses exclusively on the measure of inequity as a function of the assessed value with the house value serving as a proxy for wealth.[1] The results of previous works are often mixed and solely dependent on the value of the house to explain all the variation in assessed value, including the characteristics of the taxpaying household.[2] Further, the degree of detail has typically been limited to the entire tax jurisdiction with limited recognition of neighborhood variations (see Smith, 2002 and Birch et.al, 2004 for examples).
Utilizing the three accepted theories of property tax burden (traditional, new and benefit) this study examines the inequalities in the Illinois tax system by empirically evaluating the 2000 Illinois property tax assessment as observed in a data set of attached residential sales from the city of Chicago, Illinois. The analysis employs an expansion model similar to that developed by Casetti (1972) on a set of sales observations to test for spatial vertical inequity in the actual tax burden. In a second step, individual property tax burdens are regressed on structural and locational variables with the goal of providing expanded detail on tax inequities beyond the value of the property. The results indicate progressive inequities exist in the tax burden and the inequities exhibit an observable spatial pattern with specific characteristics of the neighborhood identified as influencing the level of inequity. The findings are timely as property taxes in Illinois and elsewhere are being challengedas the primary source of local government revenue, and state tax commissions are eager to identify tools to aid in identifying inequities and improving efficiency and quality in the assessment process.
The remainder of the paper is divided into three sections. The next section summarizes the theories on property tax burdens and how assessment of real property is conducted in Illinois. In the analytical section the data are presented, and the results of the models are discussed. The paper concludes with a summary of the findings and implications for the results.
Theories of Property Tax Burden
There are three primary schools of thought on the economic impact of the property tax. These are defined here as the “traditional” or orthodox view, the “new view,” and the “benefit” view. (Rosen, 1990, Loelli, 2001). Distinguishing between these three views is important because different classifications imply different economic effects from changes in policy, and different levels of capitalization of the property tax into ownership.
Traditional theory views the property tax as a tax on the return of investment. Under the traditional theory the property tax is divided into two components, a tax on the return on land and a tax on the return on structures. Since the amount of land is fixed, and immobile, increases in property taxes do not alter the supply of land and thereby reduce the returns to land by taking part of the revenue generated from its employ. When the property tax increases, after-tax profits decrease, lessening the value of the property that cannot be relocated. Owners pay increased taxes, receive reduced profits and own land of less value. This assumes the owners are already maximizing their profits, suggesting that if they raise their prices or increase investment in the land they will reduce after-tax income further. Since land ownership tends to increase with income, this part of the traditional theory predicts that the property tax is progressive (Aaron, 1975).
The property tax on structures under the traditional view is thought to be regressive. Given a long enough period of time, capital is mobile. When the property tax increases, barely profitable investments become unprofitable and therefore projects are not undertaken, at least not in the high-tax area. Because of the lower after-tax return to capital and the ability of investors to lower the supply of capital in the long run, firms will raise prices and pass much of the burden of the increased tax onto consumers. Since owners of structures pass the property tax burden onto consumers in this case, the tax on structures likely has regressive characteristics. The dual nature of the property tax, according to traditional theory, is that it includes a progressive tax on land and a regressive tax on structures with the regressive element dominating due to the role of improvements over land as a portion of total value (Musgraves, 1969).
In the “new view,” (see Mieszkowski, 1972) the division between land and improvements in the computation of tax burden is abandoned. Since the property tax is a tax on land, buildings, and other capital equipment, the new view looks at the property tax as a general tax on capital. Furthermore, it is assumed that capital is relatively fixed in supply. The more fixed the supply of a good, the more the burden of a tax falls on the owner/producer of that good. Capital ownership as a percentage of wealth increases with income. Therefore, the new view predicts that the tax on capital will fall more heavily on the wealthy, making the property tax a progressive tax.
The property tax is not a tax at all under the benefit theory, but is seen as a purchase price for the bundle of goods and services provided by the local government in that district such as schools, parks, police and fire protection (Tiebout, 1956). Oates (1969) provides support for this view by finding property values that increase as the quality of public services increases suggesting that homeowners are willing to pay a higher price for better public services.
The Illinois System
Real property in the state of Illinois is assessed on the basis of fair market value that is produced by application of regulations promulgated by the Illinois Department of Revenue. The property tax is the largest single tax in Illinois, and is the major source of tax revenue for local government taxing districts. The local taxing districts include counties, townships, municipalities, school districts, and special districts. Since 1930 property taxes have been levied at the local level, although the Illinois Department of Revenue issues guidelines (e.g., determining equalization factors, approving exemptions, distributing assessment manuals, etc.) and provide technical assistance to local assessing officials. The property tax cycle extends over a two-year period. CookCounty (the Chicago taxing jurisdiction) "classifies" property and assesses classes at various percentages of market value according to local ordinance. The following table of CookCounty classified assessment levels listed below encompasses over 98 percent of the total assessed value for the county.
Insert Table 1 Here
The total equalized assessed value of real property in Chicago and CookCounty was approximately $45 billion in 2002.[3] Property taxes finance a major part of the services provided by local governmental units that benefit citizens and their property. The illustration below presents a distribution of real property tax revenue as a percent of the total tax rate levied, which in 2002 amounted to 7.25% of the total equalized assessed value. In keeping with the distribution of tax revenue in many urban areas across the country, the largest share of property taxes in Chicago is allocated in support of school districts. The Chicago schools received $1.6 billion or roughly 42 percent of the $3.9 billion in property tax collected for 2002.
Insert Illustration 1 here
Although 73 percent of local government revenue is comprised of the property tax it continues to meet with persistent opposition because it often fails to satisfy a primary principle of taxation: apportion the burden of taxation on the basis of ability to pay. Another principle often invoked by framers of tax legislation is that the burden of the tax should be apportioned to those receiving the benefits of the tax funds. For a drainage district or a fire department upgrade a property tax can be rationalized with the local citizens since local fees are collected for locally provided services. The rationale supporting local taxes for public school districts is increasingly questioned, as there is little relationship between property taxes and educational benefits. This is because the benefit is increasingly exported when young adults migrate for college and employment. Local governments across the country have heard arguments in favor of altering the current assessment process because opponents view the current system as resulting in inequitable tax burden distribution between economic sectors (business, residential, industrial) and vertically based on ability to pay (Smith, 2002).
Research on the effect of the property tax on home owners typically focuses on the direct effects of increased taxes and the perceived quality of public services (see Cornia & Slade, 2005; and Smith, 2000 for surveys of previous studies). Given the property exemption structure that lessens the taxation of the poor and the concentration of the benefits of local public services at the lower end of the income scale, many property tax payers see little benefit from a property tax increase (Guilfoyle, 2000). In this case an increase in property taxes, especially for more wealthy homeowners, increases the tax burden in that locality relative to others, providing incentives for the more mobile households to relocate. The direct effect of the property tax on home owners is related to the efficiency with which the locality uses public funds, and how local public services cater to the needs of the homeowner. For example, retirees do not receive as much benefit from an increase in school funding as do parents with school age children.
The challenges are compounded by the fact that the basis of value, (the value of the real property) is constantly changing and the changes are not always homogeneous across a single, large tax area such as Chicago. The property tax requires a taxable value to be set on a specific date, but few properties sell in any given year, and property tax administrators are left to create hypothetical transactions to establish the tax base (Cornia and Slade, 2005). The resulting discrepancies create varying degrees of vertical inequity across the spatial landscape of the city. Given that neighborhoods, although difficult to define, are comprised of households that are more similar in socioeconomic characteristics than they are with households outside of their neighborhood the objective of the following analysis is to shift the way property tax inequities are viewed from a macro jurisdiction based concern to one that is segmented within a single jurisdiction. Further, the within jurisdiction inequities can be modeled as a function of representative neighborhood and structural characteristics. This extends the ability to explain inequity beyond the house value.
Testing and Illustrating the Hypotheses
The Data
The empirical study site of Chicago, Illinois represents a dynamic real estate market with expansion both outward and through in-fill development in response to strong demand and comparably aggressive overall appreciation. Further, local public officials are grappling with the design of affordable housing policies as redevelopment, and the renewed interest in the lifestyle amenities of cities (often referred to as new urbanism) have spurred rapid escalations in property rents and an associated increase in housing cost for local families.
The sample of attached single family parcels is derived from a dataset compiled by First American Real Estate Information and Services of closed property transactions. It is based on data from real estate transfer declarations filed with the Cook County Assessor between January 1992 and June 2002. The use of attached residential properties is a viable alternative to studying the traditional detached single-family market in Chicago because of the extensive number of residential units in the city that are classified as multifamily. This includes all attached residential units in the city from row houses through multistory apartments, comprising over 71 percent of the total residential housing stock in the city of Chicago (U.S. Census Bureau, 2000). The data is restricted to those properties in the classification “Single-family residences, condominiums, and apartments of 6 or fewer units” to ensure consistency in the tax rate across the sample. Properties in this classification are taxed on a basis of 16 percent of value, and properties classified as part of a development of 7 or more units are taxed on a basis of 33 percent of the assessed value.
Assessor’s records containing structural and lot characteristics, location, and assessment information are geocoded by address. The data set includes 37,556 observations coupled with information from the 1990 and 2000 U.S. Census of Population and Census TIGER FILES. The dataset originally contained 48,668 prior to reduction for missing variable observations. A difference of means test did not reveal evidence of bias in those observations that were deleted. Information from both the Census Tract and Community Area are utilized in the following analyses to serve as proxies for the neighborhood in which the individual observation is located. Community Areas in Chicago were first delineated in the late 1920s by the Social Science Research Committee at the University of Chicago. At that time they defined 75 Community Areas, using criteria designed to identify areas with separate histories and community awareness, with 2 added subsequent to accommodate annexation. The Community Area data here is aggregated from STF3 Census Tract Data. In general the first two digits of each Census Tract in Chicago refer to the Community Area number.
Test of Spatial Inequality