Property value heterogeneity and taxation

By Dave Anderson:

For every problem, there exists a simple and elegant solution which is absolutely wrong.

One of the big and stupid political food fights locally is the re-assessment of property values in Allegheny County.  These assessments were ordered by the local courts because the Pennsylvania state constitution mandates "uniform" taxation.  The judge ruled that the divergent value change rates between properties in the communities that make up Allegheny County since the last assessment led to de facto non-uniform tax rates.  Communities and home owners whose property values either lost market value or gained market value at a lower rate than county average gain rates are paying a de facto much higher tax rate on their property than individuals whose property has gained value at a faster than average rate.  Frequent reassessment minimizes the differential between de jure tax rates and actual fair market value tax rates. 

The County hired a firm to do the reassessments and the reassessment  has been a political football as it threatens to raise taxes on politically influential constituencies while lowering taxes on poor(er) communities and individuals.  The reassessment has cost the county $11 million dollars to develop the model and to get the baseline data. 

More and more people in the peanut gallery think that they have a simple, elegant and far cheaper solution.  That "solution" would be to have temps look at the recent sales prices of homes in a given community, take the price and divide the price by square footage. Do that for all recent sales, and voila, that is the average price per square foot in a borough.  Apply that number to each housing unit in the town, and all of a sudden, there is no need for a complicated forumala, expensive data collection, complex models or anything else complicated. 

The solution is simple, easy and elegant. It also makes the original injury that led to the lawsuit which has resulted in the court ordered re-assessment look like a minor flesh wound. 

Let's look at a real world situation.  The neighborhood of Garfield is in the city's East End.  It is also one of the poorest and most rapidly depopulating neighborhoods in the state.  The housing stock ranges from mediocre pre-war rowhouses, to should be condemened shells, and then Hope-6 public housing.  One can buy a small rowhouse for $10,000 (no, I am not missing a zero) in Garfield. 

At the same time, staying in the city, and just going a mile south, that same row house in the same condition will cost a buyer $175,000 in Shadyside off Ellsworth Street. The Shadyside purchaser is paying for proximity and neighborhood externalities. 

Taking an average between these two neighborhoods will lead to the Shadyside owner being assessed at slightly more than half fair market value, while the Garfield owner would see his assessment be 9 times his fair market value. 

This is an extreme example, but it shows that communities have high local variability in property values and simple solutions are not good solutions.  Complex and expensive approaches are sometimes the only way to actually make a problem tractable.