Friday, November 20, 2009

False conclusions based on flawed real estate studies

Wind energy proponents insist industrial scale wind turbines have no diminishing effect on nearby residential property values. They point to several analyses prepared in the last six years (including REPP1, Hoen2, and Hoen/Wiser3) as evidence of their claims.

These reports conclude that there is no significant relationship between distance from, or visibility of the windfarm on the sale prices of houses. However, as we've reported before, there is good reason not to place substantial weight on the findings.

The REPP report has been widely discredited due to key flaws in the methodology including the fact that sixty-six percent of the homes sampled did not see the wind facility at all and the analysis made no distinction between homes near the turbines and those five miles away thus assuming the effect of the turbines was equal on all properties regardless of proximity.

The Hoen 2006 master's thesis, while more legitimate, made clear its analysis only applied to homes near a windfarm in Fenner New York and communities similar to Fenner. Any general conclusion drawn about property impacts based on Hoen's 2006 report would be inaccurate.

And now here comes the U.S. Department of Energy's (DOE) Energy Efficiency and Renewable Energy (EERE) which has undertaken yet another study to show, once and for all, that utility-scale wind turbines do not harm property values.

In 2007, Ben Hoen teamed up with Dr. Ryan Wiser of Lawrence Berkeley National Laboratory, to expand Hoen's master's work nationwide. By June of 2007, with no report in hand, Hoen broadcast his preliminary conclusions of their study at the American Wind Energy Association's (AWEA) annual meeting. Bottom line: "No negative effects found" on property values. In November 2007, DOE touted the same preliminary conclusions at the National Association of Regulatory Utility Commissioners' annual meeting and again, Hoen appeared at the 2008 AWEA meeting where he reaffirmed his conclusions.

The Hoen/Wiser report has yet to be released, but that hasn't stopped Hoen from distributing his findings whenever and wherever possible. And since 2007, wind developers have been submitting Hoen's unpublished conclusions to State and local boards as definitive proof that properties are unharmed by the towers.

This September, Windaction.org was one of about twenty reviewers to submit comments on Hoen's work. Our full report can be viewed here. After all the hype, what we found was sobering. Windaction.org's report to Wiser and Hoen identified serious flaws in their methodology, rendering the results of their study meaningless. The independent real estate appraisal experts we consulted quickly recognized that any qualified appraiser with any experience in the regression techniques utilized -- and who was not predisposed to a preferred outcome -- would easily discredit the report.

For one, the study failed to meet a basic assumption of a regression analysis i.e. that the database of sales transactions compared were reasonably homogeneous.

The "hedonic regression method" used by Hoen argues that one can determine the marginal contribution of specific independent variables -- i.e. view of the turbines -- to the sale price. However, when variables are omitted from the model, such as number of bedrooms, or improperly weighted in assessing contribution to house price it is not possible to understand the single effect of turbine view.

Homogeneous means that the houses included in any transactions were similar in market characteristics such as approximate size, age, quality, available amenities (schools, shopping, security, access to work and recreation, etc) and were examined in a similar economic setting (employment, availability and cost of financing, market growth or decline and the like), among other factors. Homogeneity of the marketplace is fundamental to the type of regression analysis Hoen used. It is well documented that these techniques are difficult to utilize on data sets that vary substantially due to differing characteristics.

In Hoen's study, the data set spanned nine different states across the United States and included 4,895 sales of which a subset of property characteristics were identified and then averaged to produce a composite home of 47 years in age with 1,628 square feet of finished living area above ground, 1.75 bathrooms situated on 1.09 acres and having an average condition. The data showed home sale prices ranging from as low as $10,492 to as high as $647,500.

The variation in house price alone indicated a failure to meet the requirement of homogeniality. But the problem in Hoen's study was more pervasive. When we looked at the 'age' characteristic for the homes in the report, the average age of the home at the time of the sale was 47 years with a standard deviation of 36. In other words, within one standard deviation, 68.2% of the homes in the data set ranged in age from 11 years to 83 years. We have no way of knowing how age influenced sale prices within the study's data set or whether age affected house price more or less than view of the turbines. The same can be said for other property characteristics like square footage of the home, number of baths, etc. Remarkably, Hoen did not even track the number of bedrooms per house transaction or whether the houses sold had garages.

In addition, since the data set was drawn from diverse locations across the country, applying the same weight to property characteristics was inappropriate. For example, fireplaces or finished basements in Texas may be perceived as less valuable than central air conditioning and the reverse may be true for the same characteristics found in homes in upstate New York. Nonetheless, Hoen did exactly that.

Appraisers are highly critical of hedonic analysis and have warned that causal conclusions drawn about a data set when utilizing this technique are often times unsupportable. This criticism applies in cases where those conducting the analysis make every effort to obtain a homogeneous data set and include important basic property characteristics. In Hoen's case, it is clear he had no interest in conducting a legitimate study to determine the impact of wind towers on property values. On the contrary, it appears his conclusion was already predetermined back in 2007 when he raced to AWEA to promote his premature conclusions. With his "mission accomplished", unfortunately, he now has little incentive to release the final report.

1.The Effect of Wind Development on Local Property Values by the Renewable Energy Policy Project (REPP) - May, 2003

2.Impacts of Windmill Visibility on Property Values in Madison County, New York by Ben Hoen - April 2006

3.Lawrence Berkeley National Laboratory report by Ben Hoen, Ryan Wiser et. al. - Not released

No comments: