Analyzing Property Tax Appeals for Systemic Bias

About the author

Chris McLaughlin

View Other Posts

Share on Google+
Share on Reddit
Share on Tumblr

Last year I wrote about the “assessment gap” that disadvantages poor and minority homeowners in many states, including North Carolina. Intrigued by that national study of the property tax appraisal process, I decided to learn if similar evidence of systemic bias exists in North Carolina’s property tax appeal process.

My goal was to answer two questions about property tax appeal rates and results. First, do poor and minority taxpayers appeal their property tax appraisals as often as do wealthy and White taxpayers? Second, when poor and minority taxpayers appeal their tax appraisals, do they achieve similar results as do wealthy and White taxpayers?

If not, then this “appeal gap” might explain some of the previously identified assessment gap; fewer successful appeals from poor and minority taxpayers would contribute to higher tax appraisals relative to market value for those taxpayers.

Before diving into the details of my appeals study, I offer a quick note about the term “systemic bias.”  Based on reactions to my earlier post, some readers take offense at that term and the aspersions they think it casts on government employees.  But I ask those readers to remember that recognizing systemic bias in a government program is not the equivalent of accusing everyone who works in that program of being racist.  It is simply an acknowledgement that some government programs exhibit an inherent tendency to produce disadvantageous outcomes for minorities and the poor.

To see evidence of these disadvantageous outcomes in our state’s property tax system, check out this on-line tool that can produce a customized analysis of the assessment gap in (almost) any county in the United States.  (98 of North Carolina’s 100 counties were included.)

I plugged in a few North Carolina counties to see their results.  In New Hanover County, over a ten-year period the county’s most expensive homes were appraised at an average of 88% of their market value; for the least expensive homes, that figure was 103%.  In other words, expensive homes were getting a 12% discount off of market value while low-value homes were paying a small premium over market value.   According to this study, New Hanover County was the 15th least regressive county in North Carolina–or the 83rd most regressive, depending on which end of the regressivity spectrum you start from.  (Regressivity is a measure of the additional relative burden a tax system places on low-income, low-wealth taxpayers as compared to high-income, high-wealth taxpayers; more regressivity means a higher relative burden on low-income or low-wealth taxpayers.) For Durham County, the figures were 89% and 111%, making the county the 31st least regressive in the state. In Ashe County, the figures were 64% for the most expensive homes and 125% for the least expensive homes, making it the 18th most regressive county in the state (or 80th least regressive).  How does your county stack up?

Wanting to do a similar deep dive into North Carolina tax appeal data to search for evidence of systemic bias, I reached out to Dwane Brinson, the tax administrator of Durham County. He readily agreed to work with me, noting that the Durham County commissioners have made eliminating systemic bias a priority for the county.

Partnering with Dwane’s staff (shout-outs to Starlin, Kim, and Al!) and David Dunmire, a number-crunching expert here at the School, we analyzed thousands of appeals from Durham County’s last two reappraisals in 2016 and 2019 to determine if appeal rates or appeal results varied with taxpayers’ race, income, or property value.

The detailed results are in this report, but here’s the TL;DR version.  Taxpayers who owned more expensive homes and taxpayers who lived in Whiter neighborhoods generally appealed their tax appraisals at higher rates than did taxpayers who owned less expensive homes and taxpayers who lived in less-White neighborhoods.




Appeal Rates For: 2016 2019
All Properties in Study 5.4% 4.2%
High Value Properties 14.9% 11.2%
Low Value Properties 6.9% 2.8%
Low Minority % Neighborhoods 9.8% 6.2%
High Minority % Neighborhoods 3.4% 4.1%
Majority White Neighborhoods 6.6% 4.6%
Majority Minority Neighborhoods 3.6% 3.6%


While this table shows that race and property value appeared to have consistent impacts on appeal rates, those variables did not demonstrate a consistent impact on appeal results.  Appeals from taxpayers who owned less expensive homes or taxpayers who lived in neighborhoods with high minority populations generally produced similar appeal results as did appeals from taxpayers who owned more expensive homes and taxpayers who lived in neighborhoods with fewer minority residents. In other words, the average reduction in appraisal value on appeal was not dependent on property value or neighborhood racial composition.

Similarly, taxpayer income did not appear to have a consistent impact on either appeal rates or appeal results.

What do these results mean?

Appeals usually result in a reduction in tax appraisals (80% of the time, based on the data from Durham County), which means that taxpayers who appeal are more likely to have their property appraised closer to market value than taxpayers who do not appeal.   If taxpayers from Whiter, wealthier neighborhoods on average appeal their appraisals more often than do other taxpayers, then the relative tax burden on taxpayers from White and wealthy neighborhoods will be less than the burden on less-White, less-wealthy neighborhoods.  This is a clear systemic bias.

Please note that this conclusion does not mean that the officials who operate Durham County’s tax appeal process are biased or racist. The study suggests the opposite, in fact. If Durham county officials were explicitly biased against poor or minority taxpayers, we would expect to see a difference in appeal results based on property value or race. There is no evidence that taxpayers who own less expensive homes, who earn less, or who live in majority minority neighborhoods obtain worse results from their appeals than do taxpayers who own expensive homes, earn more, or who live in Whiter neighborhoods.

Assuming that explicit bias among Durham County officials is not the cause of these lower appeal rates in poorer, less-White neighborhoods, what might it be? One explanation could be that some types of taxpayers possess less information and understanding about how the appeal process works. If taxpayers with lower value homes and taxpayers from majority minority neighborhoods knew more about the process, they might file more appeals.

To combat this bias, the tax office might consider increasing its outreach to neighborhoods with less expensive homes and higher minority populations to explain that tax appeals are free, don’t require an attorney, and (usually) don’t require a physical inspection of the interior of the home.  This last point might be important for taxpayers who are concerned about inviting government officials into their homes.

Regardless of its cause, any potential bias in any government process deserves examination.  We can’t fix problems we don’t know exist.  I encourage more counties to examine their own property tax data to learn if unintended biases exist in their processes and how those biases might be mitigated.

Leave a comment ?

2 Responses to Analyzing Property Tax Appeals for Systemic Bias

  1. Witt Putney says:

    This is an interesting study. I do have a few points I feel are worth mentioning….

    1. How did the study account for owner occupancy vs. rental property? My assumption is that lower value neighborhoods have a higher percentage of rental properties. The Census would include data on the inhabitants of the property, but the inhabitants may not have appeal rights if they rent. Would the analysis have changed if the data was limited to properties where the mailing address for the owner was the same as the situs address of the property?

    2. An appraisal is an opinion of value, with the “appraised value” often formed from a reconciliation of values from multiple appraisal methods. Properties that have higher valuations typically have wider ranges of indicated values that must be reconciled. I liken this to the upward bias evident in price-related differential (PRD) analysis where the acceptable range does not evenly deviate from 1 but instead ranges from .98 to 1.03. As the value gets higher, I believe there is more “wiggle room” on what constitutes market value. A lot of those homeowners are savvy enough to recognize this divergence and the potential tax savings are high enough for them to take a shot at an appeal.

    3. Higher value homes tend to have features that are difficult to assess in mass appraisal. Outdoor kitchens, palatial basements finished out to a level on par with above-grade levels, and the surge of solar panels and green construction standards are just a few examples of amenities that are difficult to capture in mass appraisal models.

    These points aren’t meant to refute your findings. I simply bring them up as potential additional reasons for why assessment ratios may skew, particularly when viewing the data across the market value spectrum.

    • Witt, thanks for your thoughtful comments. As to #1, you correctly point out that the study included rental properties. We did not limit the study to owner-occupied housing because residents of rental properties suffer from an assessment/appraisal gap just as do owner-occupiers; the resulting higher-than-appropriate appraisals are passed along indirectly to the tenants thru higher rents.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.