By Yulong Chen, Liyuan Ma, Zachary Martin, and Peter F. Orazem
Housing wealth represents about half of the wealth held by US households (Iacoviello 2011) and is the primary asset for most households. Housing represents 80% of the wealth for the bottom half of the wealth distribution, two-thirds of the wealth for households between 50%–90% of the wealth distribution, but only a small fraction of the wealth of the top 10% of households (Kuhn et al. 2020). Consequently, households at the bottom of the wealth distribution are much more exposed to changes in housing values. During the run-up in housing values between 1971–2007, wealth rose fastest for the bottom half of the wealth distribution. While all assets fell in value during the Great Recession, the slow recovery of housing prices compared to stock prices contributed to rising wealth inequality in the United States.
Because homeownership represents the dominant source of wealth for most households, government policies that alter housing prices can have important consequences for household asset accumulation. Government services such as education, public safety, roads, or parks can raise the value of living in the community. Households choose where to live based on their preferences for local government services (Tiebout 1956). On the other hand, local taxes are required to pay for government services, and the taxes lower the value of home ownership. Oates (1969) argues that the costs and benefits of residing in a location would be capitalized in property values with public goods raising housing prices and property taxes lowering those prices. It is possible that the tax and government service effects counteract one another exactly, in which case the often substantial variation in local government expenditures and taxes have no impact on housing values or household asset accumulation.
We examine the extent to which housing values capitalize on the effects of several location-specific taxes and services including property, sales, and income taxes and education expenditures. We have an additional interest in examining whether the effects differ in rural areas that have lower property values on average. Rural housing values average 42% lower than urban housing values, and rural incomes lag urban incomes by 15%. As a result, rural communities need to apply higher tax rates than urban communities to generate the same per capita tax revenue.
One concern is that these government policies may be altered with the aim of altering their effects on housing values, in which case the relationships may not be causal but correlational. A commonly used statistical solution is to exploit the often-large tax rate and government expenditure differences that occur at state borders. Households that have decided to reside in the area can decide which side of the state border to live on based on their preferences for government services and tax policies. This discontinuity in government policies at the border resolves the endogeneity problem.
Analysis
We conduct our analysis using county-level data over the period of 2000–2017. The model represents a regression of the difference in housing prices across adjacent counties at a state border on differences between the counties in tax rates and government expenditures between the states. Our measure of housing values is the median value of housing prices in each county in 2017 dollars. We add controls for the age and square footage of the house to adjust for differences in housing quality. Our tax rate measures include the state sales and income tax rates and the effective property tax rate in the largest city in the state. We add an indicator variable if the state has a tax reciprocity agreement with its neighboring state.
Our measure of local government service is the per capita expenditure on education in 2017 dollars. We add a control for differences in naturally occurring county amenities that could plausibly affect the incentives to live on one side of the border versus the other. Differences in county per capita incomes would influence the ability to pay and housing demand.
There are reasons to suspect that differences in income tax rates and sales tax rates will not affect housing values at state borders. Because individuals can easily work on either side of the border, workers will have an incentive to take jobs in the lower income tax state. Firms on either side of the border will compete for workers by offering the same after-tax wage, and so the higher income tax state will pay higher wages, neutralizing the adverse effect of the income tax. Consumers can shop on either side of the state border, and so they will have an incentive to shop at stores facing lower sales taxes. Retailers on the higher sales tax side of the border will have to lower their prices to compete, and so the after-tax price of goods will equalize, neutralizing the adverse effect of the sales tax.
Land cannot migrate in search of the lower property tax rate. As a result, housing values must fall on the side of the border with the higher property tax rates to equalize after-tax housing costs. Consequently, we would expect that with competitive labor, goods, and housing markets, only the property tax rates will be capitalized into housing prices.
Empirical results
Our regression analysis strongly confirms that the property tax is capitalized into housing prices. An increase in property taxes of one mill rate ($1 per $1000 in property value) lowers housing values by $199 or 0.16%. The average property tax rate across the 48 contiguous states in 2017 was a mill rate of 16.6, meaning that the average loss of housing value due to the imposition of property taxes is $3301. This loss of wealth is disproportionately borne by households in the lowest 90% of the wealth distribution, the households whose wealth is concentrated in housing.
The other taxes have no effect on housing values at state borders. Because households can react to income taxes and sales taxes by choosing which side of the border to shop or work, wages and prices adjust so that after-tax wages and retail prices equalize across the state border. The high-income tax state pays a higher wage and the high sales tax state charges lower retail prices.
Of the other measures that could plausibly influence housing prices, localities with higher per capita incomes have higher housing values. Localities with superior natural amenities also have higher housing values. However, there is no significant effect of higher local spending on public services, such as that of education spending on housing prices.
We had expected that the property tax might have a different impact on urban versus rural property values. In fact, the effects of tax rates, government expenditures, per capita incomes, and natural amenities were virtually identical. As a result, property taxes do not alter the relative wealth of urban versus rural dwellers.
Property taxes do affect states differently, however. In table 1, we provide a ranking of states based on how greatly their property tax structure affects housing values in the state. We apply the average mill rate effect of -0.16% per mill rate to the 2018 mill rate charged in the largest city in each state. Colorado has the least distortionary property tax effect of -0.9% reduced property values because of their property tax levy. At the other extreme, Michigan households sacrifice 5.7% of their housing value due to their property tax levy. The states that lose the greatest wealth due to their use of the property tax are Michigan, Connecticut, Indiana, New Jersey, and Wisconsin.
| Rank | State | Tax Effect | Rank | State | Tax Effect |
|---|---|---|---|---|---|
| 1 | Michigan | -5.7% | 25 | Missouri | -2.2% |
| 2 | Connecticut | -5.2% | 26 | Pennsylvania | -2.1% |
| 3 | Indiana | -5.0% | 27 | South Dakota | -2.1% |
| 4 | New Jersey | -4.8% | 28 | Kentucky | -2.0% |
| 5 | Wisconsin | -4.7% | 29 | Oklahoma | -2.0% |
| 6 | Iowa | -4.0% | 30 | Minnesota | -2.0% |
| 7 | Texas | -4.0% | 31 | Kansas | -2.0% |
| 8 | Maryland | -3.6% | 32 | California | -1.9% |
| 9 | Vermont | -3.4% | 33 | North Dakota | -1.9% |
| 10 | New Hampshire | -3.4% | 34 | South Carolina | -1.9% |
| 11 | Nebraska | -3.3% | 35 | Nevada | -1.8% |
| 12 | Maine | -3.3% | 36 | Oregon | -1.8% |
| 13 | Illinois | -3.2% | 37 | North Carolina | -1.6% |
| 14 | Ohio | -3.1% | 38 | Wyoming | -1.6% |
| 15 | Mississippi | -3.0% | 39 | Massachusetts | -1.5% |
| 16 | Rhode Island | -3.0% | 40 | Montana | -1.5% |
| 17 | Florida | -2.9% | 41 | Washington | -1.5% |
| 18 | Georgia | -2.8% | 42 | Virginia | -1.4% |
| 19 | Delaware | -2.7% | 43 | Tennessee | -1.4% |
| 20 | New Mexico | -2.5% | 44 | West Virginia | -1.4% |
| 21 | Louisiana | -2.4% | 45 | New York | -1.3% |
| 22 | Idaho | -2.3% | 46 | Utah | -1.2% |
| 23 | Arizona | -2.3% | 47 | Alabama | -1.1% |
| 24 | Arkansas | -2.2% | 48 | Colorado | -0.9% |
Conclusion
Every state uses property taxes to pay for local public goods. The long-standing expectation was that better public services raise property values while taxes lower property values. However, we find that the negative effects of property tax rates outweigh the benefits of the public goods that they finance. Other taxes, such as sales tax or income tax, have less distortionary effects on property values. Because the least wealthy households have a greater share of their wealth in housing, the property tax places a greater burden in the form of lost wealth on the bottom 90% of the wealth distribution. However, the effects are nearly identical in urban and rural counties, and so the property tax does not create a source of wealth inequality between urban and rural residents.
In another context, Chen et al. (2022) find that property taxes were also the most important in reducing the rate of new firm entry. Because entrepreneurial ventures are another mechanism for building wealth, it is interesting that property taxes limit asset accumulation in that context as well.
Acknowledgments
Corresponding author: Peter F. Orazem, University Professor of Economics and Director, Program for the Study of Midwest Markets and Entrepreneurship, 267 Heady Hall, Iowa State University, 518 Farm House Lane, Ames, Iowa 50011-1054, (515) 294-8656, pfo@iastate.edu. We are grateful for partial research support under USDA-NIFA grant 2018-68006-27639 and a grant from the Charles Koch Foundation.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
Chen, Y., K. Duncan, L. Ma, and P.F. Orazem. 2022. “How Relative Marginal Tax Rates Affect Establishment Entry at State Borders.” Small Business Economics, Forthcoming.
Iacoviello, M. 2011. "Housing Wealth and Consumption." Federal Reserve Board International Finance Discussion Paper 1027.
Kuhn, M., M. Schularick, and U.I. Steins. 2020. “Income and Wealth Inequality in America, 1949-2016.” Journal of Political Economy 128(9): 3469–3519.
Oates, W.E. 1969. "The Effects of Property Taxes and Local Public Spending on Property Values: An Empirical Study of Tax Capitalization and the Tiebout Hypothesis." Journal of Political Economy 77(6): 957–971.
Tiebout, C.M. 1956. "A Pure Theory of Local Expenditures." Journal of Political Economy 64(5): 416–424.
Suggested citation
Chen, Y., L. Ma, Z. Martin, and P.F. Orazem. 2023. "Property Taxes and Housing Prices in Urban and Rural Markets." Agricultural Policy Review Spring 2023. Center for Agricultural and Rural Development, Iowa State University.