I have received several inquiries regarding the “new” Ben Hoen study on residential property values. It appears to be just a regurgitated version of the December 2009 study. So it is the same “data” he relies upon, and nothing is new except some different spin.
I have developed a few observations and rebuttal comments for consideration by anyone interested in wind turbine value impacts, and my opinions as to the reliability of the linked paper and underlying data are as follows:
1 – Hoen Excludes Relevant Data – Hoen claims to use about 7,500 sales, yet he excludes certain sale data from nearby turbines, i.e., developer buyout and resales at 36% & 80% reduction upon resale. Had just those two (2) sales been included, it would impact his “Nuisance” effect from about 5.3%-5.6% up closer to 10%. See figure ES-1, page xiii, in original Dec. 2009 report. Note that Figure ES-1 is conveniently excluded in this newer paper. Perhaps the negative finding, even though claimed to be statistically insignificant, does not help the cause of getting turbine projects approved? Further, see footnote 27 in the Dec. 2009 report on pages 13 & 14. Hoen claims that of the four (4) sales to developer that resold, two (2) were to a “related party”. They were in fact sold to parties who were unrelated to the developer, and his exclusion basis is misleading, and is contrary to the website “assessor manual” description of invalid sales. In short, it would seem he deviated from established rules to exclude that inconvenient data. The ES-1 data results also do not reflect the fact that two (2) of the buyout properties apparently could NOT be resold. Finally, he excludes 34 sales that were coded by the counties as “valid” sales (i.e., arm’s length, no foreclosure or estate sale, etc.) since they had resold, without any explanation if they resold due to turbine nuisances, or if they reflected discounting or appreciation between sales. He also excluded 5 sales that were more than six standard deviations from the mean. Pardon my bluntness, but if one is looking for an honest answer to turbine impacts, the excluded sales are exactly where the focus should begin, since they show considerable probability of having been impacted by the nearby turbines.
2 – Hoen Opinions Are Misquoted in a Misleading Manner – He does NOT say there is no impact on value. He DOES opine that there is no “statistically significant impact on sale prices”. First sub-point: When the inventory of houses available IS significantly impacted by a large scale disamenity via acting as a deterrent to buyers, then often homes will sit on the market for extensive periods of time, usually referred to by Realtors as Days On Market (DOM). Numerous examples of unsold inventory demonstrate that turbines have exactly this effect on the majority of would be buyers, when the turbine presence is “dominant” or spoils natural view amenities. Examples of homes that sold after protracted marketing range from 20% to 40% value reduction after 2-3 years on the market, and further, the developer resales in the preceding Point 1 demonstrate that in order to obtain a more “typical” DOM, the prices need to be discounted even steeper…as much as 80%. Those discounted prices ARE value by definition, since that is what a willing buyer paid to a willing seller, neither being under compulsion to consumate the transactions. Thus, Hoen uses the absence of sale evidence to claim “prices” are not impacted in a statistically significant manner. Second sub-point: In his 2006 thesis, which was a precursor to the USDOE funded study, Hoen made similar claims. But if one examines the map in the attached Fenner study (page 47 in report; pdf page 58 of 73), it becomes apparent that there is an absolute dearth of sale activity within about 3/4 to 1 mile of the turbine cluster in the Fenner project. I call this the “doughnut hole” effect, since if you stare at that map long enough the hole in the data becomes obvious. Third sub-point: In his 2006 thesis, Hoen states that data for marketing time (DOM) was not available (report page 23). Yet, he has now had 6 years to thoroughly research and analyze DOM data, which can be printed out by any local Realtor via MLS statistics. Despite at least 2 new publications since his thesis, he has yet to report any MLS marketing time statistics. Also see last few lines of page 31 and first several lines on page 32: Hoen acknowledges that the Fenner area home values have decreased, which is confirmed by a local Realtor, but by adjusting the Township variable, his application of regression number crunching is able to make that effect go away, from his statistical perspective (See “Rubber Rulers”, below). Also, see page 42, where Hoen admits “second homes” (i.e., Nantucket, Cape Vincent, Wolf Island, etc.) are not analyzed at all. This exclusion carries forward through to the 2009 LBNL report and the latest version of that report currently making the rounds.
3 – Hoen’s Methodology is Unreliable – First, see “Wind Farms, Residential Property Values, and Rubber Rulers”. Al Wilson is a seasoned (now retired) and highly qualified real estate appraiser, and an expert on the use of regression analysis for mass appraisal purposes. He reviewed the 2009 Hoen/LBNL report and concluded that “the report should not be given any serious consideration for any policy purpose. The underlying analytical methods cannot be shown to be reliable or accurate”. I will add that the most authoritative appraisal text on evaluating detrimental conditions lists the 3 basic methods of determining property value damages from a neighboring cause, in descending order of reliability, as follows: 1) Case studies. The developer resales cited in Point 1 are a good example of case study type of data. 2) Paired Sales. See McCann Mendota Hills study – sales near (<2 miles) vs far (>2 miles) from turbines show a proximity effect averaging 25%. 3) Regression analysis. Problems with Hoen’s use of this overall method and unproven model are stated clearly in the Rubber Rulers paper by Al Wilson. How statistics can be used to mislead is summarized clearly in the article by Dr. Alec Salt: “Why do pro wind studies often use a 10 km radius?” Finally, Hoen prepared the report for a branch of the Federal Government (USDOE), and the report is certainly a public document. Nevertheless, Hoen has reportedly been steadfast in his refusal to provide the raw sale data to anyone, including the people he claims did a “peer review” of his report. I know this for a fact, because I was an invited “peer reviewer,” and he refused to provide the data to me to facilitate the requested peer review. Keep in mind that the data was from various Counties throughout the USA, and that data is public record. However, Hoen claimed it was obtained subject to confidentiality agreements. This does not make sense, and is far from any kind of scientific peer review basis supporting his opinions. Thus, the claim of “peer review” is also misleading, since scientific peer review includes the ability to review the author’s data details to test the veracity of the conclusions.
4 – Hoen (et al) Not Licensed as Appraiser – Ben Hoen has had absolutely no professional real estate experience, and he holds no license to practice as an appraiser. To my knowledge, he has never been qualified in any court as a “property value” or “appraisal” expert, and reliance on his opinions is potentially a negligent act, similar to taking legal advice from a non-lawyer or medical advice from a student. With the public at considerable risk of property value and equity losses, a generously compensated pro-wind advocate is the least reliable source for independent, professional advice or “expert” opinion publications upon which to base far reaching land use approval or policy decisions. Page x of the 2009 report states: “It should be emphasized that the hedonic model is not typically designed to appraise properties …”. However, The Hoen report is often used by developers to assert a claim that there is no impact on value from wind turbines which is, in fact, a value opinion.
5 – Hoen Recommendation for PVG’s is Contrary to Claim of No Impact on Value – See Hoen’s May 2010 webinar slide 32, suggesting that PVGs should be used to mitigate value loss risks. Also see written & recorded comments from telephone interview of Ben Hoen by Clif Schneider, where Hoen admits developers misquote his opinions and report. Finally, a AWEA 2011 policy report states they are working on developing a PVG solution for wind project developers. One does not need to be an appraiser to understand the contradiction of recommending value guarantees if there are “zero” impacts on residential values.
6 – Hoen Admits Little Knowledge of Impacts Within 1/2 Mile of Turbines – Hoen gave a presentation at the March 7, 2012, AWEA Midwest Summit. See slide 14 – suggests more research needed. Note: Per slide 4, he states his 2009 study indicates “Lack of consistent evidence of post construction effects based on distance from or view of turbines in all models”. He does not opine “NO impact”.
I trust this critique will assist the reader in better understanding the risks inherent in relying upon the Hoen reports for any land use or policy decisions.
Michael S. McCann, CRA
McCann Appraisal, LLC
This article is the work of the author(s) indicated. Any opinions expressed in it are not necessarily those of National Wind Watch.