Damned If They Do . . .

A month ago, when I commented on Pittsburgh's property reassessment controversy, there was some evidence from a small sample of three streets that the new values still favored the wealthy and discriminated against the poor. Rich Lord of the Pittsburgh Post-Gazette has done a more complete analysis of the reassessments, and the results are as I suspected.

For non-Pittsburghers, let me back up a step. Allegheny County property had not been reassessed since 2002. Failure to do regular reassessments discriminates against low income residents, whose property values are usually stable or declining, while giving a tax break to higher income residents living in areas where property values are on the rise. A coalition of low income homeowners filed suit to force a reasseesment and won. Judge Stanton Wettick ordered that 2012 taxes to be based on the new values. These new values were sent to Pittsburgh homeowners in December, causing much wailing and gnashing of teeth. When Allegheny County Executive Rich Fitzgerald argued that there was not enough time for the appeal process to run its course, Wettick backed down and gave county residents another year of the old values.

The problem is that the new values are also regressive. A regressive tax is one in which the tax rate goes down as the real value of the property subject to taxation increases. Therefore, the burden of taxation falls more heavily on the poor than the rich. Lord found that the reassessment systematically overestimated the value of low-priced land and buildings, while underestimating the value of higher-priced properties.

Lord analyzed 2099 “arms-length” sales that took place in 2011. An arms-length sale is one in which the buyer and seller have no prior relationship to one another. If a prior relationship exists, as when a home is sold to a relative, the sale price may not reflect the true market value of the property. The reassessed values were divided by the sale prices of these same properties to yield a percentage. So, for example, a score of +50% means that the reassessed value of the property is 50% higher than the sale price, as when a home that sold for $50,000 is assessed at $75,000. Here are the percent differences, aggregated by to the sale prices of the properties.

Sale price	% difference between assessment and sale price
$5,000 to $19,999	+295%
$20,000 to $39,999	+134%
$40,000 to $74,999	+56%
$75,000 to $99,999	+17%
$100,000 to $149,999	-3%
$150,000 to $249,999	-17%
$250,000 to $999, 999	-31%

As Lord points out, the houses that sold for between $100,000 and $150,000 are pretty accurately assessed, but the more valuable properties are underassessed. The greatest unfairness, however, occurs among the low-value properties, which are assessed at much more than they are worth. However, an undervalue of 30% on a $500,000 property is worth a lot more money to the owner and a lot more tax revenue to the county than an overvalue of 30% on a $50,000 property.

Click here for a more complete table, showing how these percentages differ by wards within the city. In the interest of full disclosure, I should note that I live in Highland Park (Ward 11), and as would be predicted from the table, our home is assessed at less than we paid for it in 2009.

The reaction to the P-G study of Wesley Graham, the county's acting chief assessment officer, is instructive. He complained that they had not verified that all 2099 sales were between independent parties. Presumably, there could have been some undetected sweetheart deals among the properties with lower sale prices. I must confess, I'm also concerned about including extremely low-end properties in the analysis. Some sellers may have been trying to unload run-down properties by selling them at below market value. However, even if we disregard the top two rows of the table, the figures are still quite persuasive.

Mr. Graham also said that the reassessment values meet industry standards for fairness and consistency. It's not clear from the article what percentage of deviation from the sale prices the industry considers acceptable. Regressive property taxes may be more common than is generally acknowledged. Many years ago, I had a sociology professor who told our class that tax assessment was one of the most important but overlooked ways in which poor people get screwed in this country. He argued that tax assessments are almost always regressive, or as he put it, the assessment curve is always flatter than the curve describing the real value of the properties. (I remember this well because I had a summer job working for a tax assessor at the time.) The people who profit from this are not only people with expensive homes, but also the owners of commercial property, which is almost always undervalued. Does anyone know of any good national studies of this problem?

Why are property values almost always regressive? Dominic Gambino, former manager of the Office of Property Assessments, said, “It's very hard to catch those quickly appreciating properties.” He didn't explain why the P-G could catch them while the county couldn't. One of Mr. Graham's comments may have been closer to the truth. He said assessors are “a little gun-shy about putting on the value that they're seeing” on high-end properties. Could it be that they know that affluent people are more likely to appeal their assessments than poorer people?

The Post-Gazette also reported some anecdotal evidence that suburban assessments, which have not yet been posted on the web, are also regressive. For example, property values in low income communities like Braddock and Duquesne appear to have increased more than those in wealthier suburbs like Monroeville and Plum.

Chris Briem has pointed out that, in spite of their bias, the new values are more accurate than the 2002 figures. Whether the county uses the old assessments or the new ones, however, it looks like the poor will always be shortchanged.