Two-Rate Property Taxes on Construction
In collaboration with Tyler Piddington and Maxx Marcus
Taxes have, for a long while, been a hotly contested subject in the United States and all over the rest of the world. The premise is simple, take some money from everyone in a society and use it to provide services that are better provided by the government than by private enterprise. A good idea on the face, but how exactly should it operate? Who should pay what amount? What is this amount based on: income, wealth, the value of property owned? What should these funds support: poverty relief, housing inadequacies, infrastructure, health care, public safety? Politicians and the general public constantly debate these questions. Some argue in favor of a progressive tax, that the wealthy should pay more as they possess more and it is less of a burden on them. Others argue a flat tax is fairer; everyone pays their “fair share” of the burden. The uses of the proceeds of these tax dollars is an equally debated topic, but one that usual centers around personal beliefs.
The main goal of Plassman and Tideman (2011) is to evaluate the effectiveness and impact of a two-rate property tax system, particularly in light of what happens to construction activity. A two-rate tax system is one where land is taxed at a higher rate than improvements. In all 50 states, except Pennsylvania, land and improvements are taxed at the same rate. The hypothesis put forth by Plassman and Tideman, is that if a two-rate system was implemented, construction activity would increase. Within this overarching goal, the authors also examine and analyze previous studies which did not show a statistically significant impact of the two-rate taxes in Pennsylvania in an attempt to determine why they reached the conclusions they did and to find ways to appropriately adjust the data to achieve a statistically significant result.
In a general sense, Plassman and Tideman (2010) use construction data from Pennsylvania to show that cities that have adopted a two-rate tax experience higher level of construction activity than they would if they had a one-rate tax in place. What makes the specific data used in this study so critical is that there are peculiarities in the data that make the distributional assumptions used of great importance. The reason for this is that there are years where municipalities don’t experience any new construction, so this places a large value at zero, which skews the distribution. In terms of which construction permits are useful to the study, the Bureau of the Census has created 21 different categories of building permits. Of those categories, 18 refer to new construction, while 3 consist of additions and alterations to existing buildings (not new construction of whole units). For the purpose of the study, the authors omitted permits from these three categories. The authors then ran a number of tests based on population trends to eliminate municipalities that were significantly different. After this, the authors had 219 municipalities, of which 15 had adopted a two-rate tax.
The main goal of the paper is to determine whether or not a two-rate tax has any effect of the rate of new construction. This appears to be a complicated question to answer and one that is far from straightforward to answer. After careful consideration, Plassman and Tideman (2010) address the distributional problem by first analyzing the effect of two-rate taxes on the number of building permits, and then their effect on the value per permit; the product of the two estimates yields an estimate of the effect of two-rate taxes on the total value of construction. In order to do this kind of analysis, the authors needed to separate the tax impact on the number of building permits and on the value per permit for four different categories of construction (residential whole units, residential additions and alterations, nonresidential whole units, and nonresidential additions and alterations). As far as economic intuition, their method makes sense in terms of finding the value of construction through the number of permits multiplied by the value of the permits and making a comparison across one-rate and two-rate municipalities. A large part of Plassman and Tideman’s contribution to this topic is their use of a Poisson distribution instead of the standard distributions used by previous academics. The Poisson distribution is positively skewed and so it is much better suited for the type of data (significant number of zeros) that this research yielded. The Poisson distribution gave their results much more statistical significance than previous studies that used standard regression models.
The authors found that previous analyses of the impact of two-rate taxes used standard regression techniques and focused on the value of construction. The standard regression models did not fit the data well, thus many previous findings were not statistically significant. What makes Plassman and Tiderman’s research different is that they examined separately the tax impact on the number of building permits and on the value per permit. By separating out the tax effect on number and value of permits and by using a Poisson distribution, they were able to find a positive and statistically significant impact on the number of permits. The estimated overall impact on the total value of construction is positive and statistically significant.
In conclusion, there is a statistically significant positive impact on construction activity when a two-rate tax is imposed. This is logical because if the value of improvements is taxed at a lower rate than land, there is incentive to develop the land into something more economically productive. In the case where land and buildings are taxed at the same rate, construction is discouraged as it increases the tax burden on the property owner relative to the two-rate system. There is evidence that it does increase construction; some would argue this is good for society, whereas others would argue the opposite. A lower rate on improvements also raises the question of where that shortfall in tax revenue would come from; which of course lends itself to a multitude of options, none of which everyone would agree on. While Plassman and Tideman conclude a two-rate tax leads to more construction, the best use of this information is another question.
If the authors were to do further research on this topic, it would be interesting to see what the effects of the two-rate tax are in other areas besides construction activity. For example, are other tax rates higher to make up for the shortfall from taxing structures at a lower rate? Are cities with two-rates even able to make up for the shortfall or do they experience frequent budget deficits? Additional research could also consist of surveys to developers and other players in the market to see how they view the two different taxing systems and any effects the difference has in the eyes of a developer. Additional research could also include looking internationally at what other countries do when it comes to taxing real estate. Do they even tax it? Do they use a one- or two-rate system? The answers to these questions would provide valuable insight into the desirability of a two-rate system. Perhaps more states/municipalities will adopt this system in the future if it is believed that the benefits it provides by stimulating construction offset the revenues lost from a lower rate on structures.
Multiple Choice Questions
-
1. According to Plassman and Tideman (2010), what was the impact of the two-rate tax system based on the total value of construction?
a. The estimated overall impact on the total value of construction is positive and statistically significant
b. The estimated overall impact on the total value of construction is positive and statistically insignificant
c. The estimated overall impact on the total value of construction is negative and statistically insignificant
d. The estimated overall impact on the total value of construction is negative and statistically significant
Explanation: Based on the findings of Plassman and Tideman’s study the impact of the two-rate tax system based on the total value of construction is positive and statistically significant. Also the impact was found to be positive and statistically significant based on the number of permits, but statistically insignificant for the value per individual permit.
-
2. According to the Building Categories in the Bureau of the Census Data Set, which of the following selections are actual building permit categories?
a. Single-family houses
b. Addition of garages and carports
c. Three-and four-family buildings
d. Industrial buildings
e. Barns
Explanation: After reviewing Plassman and Tideman (2010); results and viewing the building categories, one can clearly pick out answers (a)–(d) and see that they are all actual building permit categories of both the residential housekeeping buildings as well as the residential nonhousekeeping buildings and nonresidential buildings. Answer (e) was incorrect as a barn is not its own category but would fall under a nonresidential structure other than a building.
-
3. What assumption is made under the Poisson model and why is this essential to the study of Plassman and Tideman (2010)?
a. The data have a variance that is equal to their mean; it’s essential to this study because the distribution is meant to model the number of building permits issued, a number that can’t be negative and will be positively skewed as it is typical for only a small number of permits to be issued in a given municipality in a given year
b. The data have a variance that is greater than their mean; it’s essential to this study because the distribution is meant to model the number of building permits issued, a number that varies greatly and will often be positively skewed as the number of permits issued in a given municipality in a given year is highly unpredictable
c. The data have a variance that is less than their mean; it’s essential to this study because the distribution is meant to model the number of building permits issued, a number that doesn’t vary in a given municipality in a given year
d. There are no assumptions when using the Poisson Model
Explanation: In this study the answer is (a), based on the findings of both Plassman and Tideman (2010) and the needed use of the Poisson Model. It is essential to the study because it is meant to model the number of building permits issued. In addition, we do not believe that the observed request for buildings is best conveyed as a distinct outcome of an otherwise unceasing choice process, but that the Poisson process defines the demand for such structures more accurately.
References
Plassman, F. and T. N. Tideman (2010), “A Markov Chain Monte Carlo Analysis of the Effect of Two-Rate Property Taxes on Construction,” Journal of Urban Economics 47, 216–247.
Plassmann, F. and T. N. Tideman (2011), "Marginal Cost Pricing and Eminent Domain,” Foundations and Trends in Microeconomics 7, No. 1 1–110.