Yet another source of inequality: Property taxes

Schools, roads, and fire and police services—all are financed, in large part, by property taxes. These taxes are a pain to pay, bureaucratically dense, and absolutely essential.

They’re also an entrenched source of racial inequality.

Recent analysis from the Minneapolis Fed shows that Black and Hispanic homeowners face property tax assessment rates between 10 percent and 13 percent higher than White homeowners in the same tax jurisdiction. That translates to an additional tax burden of $300-$390 per year, on average, for Black and Hispanic homeowners, creating a significant barrier to wealth accumulation.

In “The Assessment Gap: Racial Inequalities in Property Taxation,” a working paper from the Opportunity & Inclusive Growth Institute, visiting scholar Troup Howard of the University of Utah, with Carlos Avenancio-León of Indiana University, describes two factors that generate this assessment gap—spatial or “neighborhood” effects, and nonspatial or “individual” effects. And the economists then identify and measure a key force that operates at each level.

At the neighborhood level, they demonstrate, faulty valuation of amenities systematically boosts tax assessments on minority-owned homes relative to their market value while lowering nonminority assessments. At the individual level, the economists document, minority owners are less likely to appeal assessments than nonminorities. When they do appeal, they tend to win less often, the data show, and if they win, they win less financial relief than nonminority owners.

Fortunately, Avenancio-León and Howard also find a practical solution to this institutionalized inequality. Linking assessment growth to ZIP-code-level housing price indexes would substantially reduce racial bias in tax assessment. Using price indexes at finer geographic calibrations would decrease it still more.

Home bias

Virtually every U.S. state constitution or legislative code provides that property taxes are to be ad valorem—proportional to a home’s fair market value. Local officials are entrusted with annually assessing home value as a reflection of the quality of the structure itself (square footage, number of bathrooms, and so on) and the appeal of its location (the cardinal rule of real estate).

The economists’ central insight is that while markets do a good job of evaluating both home quality and location, government assessments are insensitive to the latter. Assessment ratios cover broad geographic areas, and amenities that matter significantly to homeowners vary widely within them. The result: A four-bedroom house located on a busy street might be assessed at the same rate as a physically identical house beside a quiet meadow.

By itself, this generates assessment-value mismatch, but not systematic bias. But as the economists point out, residential spatial sorting by race is a well-established fact in the United States, and minorities tend to live in lower-amenity neighborhoods. So the owner of that hypothetical home on the busy street is likely to be minority, and the meadow-dweller, probably White. Assessors don’t sufficiently take amenities into account, the data show, so they routinely underassess the average White-owned home and overassess the average minority-owned home.

This neighborhood composition channel accounts for a large part of the total assessment gap, but not all of it. Even after controlling for spatial factors by looking within small geographic areas like U.S. Census blocks, the economists find that the average minority homeowner faces an assessment 5 percent to 6 percent higher than his or her nonminority neighbor. “This latter finding is particularly surprising,” they note, “given that most assessors likely neither know, nor observe, homeowner race.”

After careful analysis of just a single U.S. county (albeit the nation’s second largest—Cook County in Illinois) and verifying against national assessment patterns, they establish that minority homeowners are less likely than Whites to file an appeal of a tax assessment, less likely to win appeals, and less likely to win as large an appeal as White owners of comparable-value homes.

A thicket

To arrive at these disturbing findings, the economists first painstakingly hack through the thicket of U.S. property tax administration. More than 75,000 government entities potentially impose property taxes on U.S. homeowners, and every home faces taxation from multiple units. Cities and counties have a hand in the pot, of course, but so do school districts, park boards, municipal utility districts, and the occasional “special purpose district” (airport authorities and regional economic development initiatives, for example). Geographic boundaries of these taxing entities rarely coincide. As a result, the average U.S. home is subject to property taxes from 4.5 local units.

To measure a potential assessment gap, Avenancio-León and Howard need to compare two residents for whom the tax burden should be identical. That means the homeowner should be served by the same set of government units, receive the same public goods such as schooling and street maintenance, and face the same tax rate. To make that comparison, the economists establish exactly which taxing entities touch every single property in their sample.

By design, they limit their sample to homes sold in an “arm’s length” transaction—meaning both buyer and seller are independent, with equal bargaining power. That sale price establishes fair market value. They also record the home’s tax assessment from that same tax year (ensuring that the assessment is prior to sale so that sale price doesn’t affect the assessment). From government data, they collect the racial and ethnic identity of mortgage holders. And from the American Community Survey, they merge in key information on neighborhood attributes.

In the end, the economists create a panel of 6.9 million homes across the country, each bought and sold at least once between 2005 and 2016. It’s a broad sample with all the elements needed to measure the gap: every home’s patchwork of taxing units, annual assessment figure, recent transacted sales price, and race of homeowner.

Conclusions and solutions

The results summarized here are a fraction of the total. Other findings:

• States vary widely—Illinois, Missouri, and Ohio have the largest assessment gaps; Vermont and Indiana have low or “negative” gaps.
• The assessment gap is larger in areas with higher shares of minorities and for lower-income owners. (See Figures 1 and 2.)
• The gap is also larger in areas where residents have high racial prejudice. The economists suggest that this may be due to higher racial residential sorting.

There is some good news: The economists demonstrate a practical approach for addressing the pattern of distortion between market prices and assessment ratios. When a home sells, set that sales price as the assessment of fair market value. Then index assessment growth to a housing price index for the smallest geographic area available.

Using a ZIP-code-level price index would trim the overall assessment gap by more than half, they calculate, and a more refined index could do far better. Such a rules-based approach, note Avenancio-León and Howard, would also increase transparency in the assessment appeals process.