leverage expertise in various practice
areas, not merely those of an appraiser
or economist.
Recognizing these shortcomings, we
have been compiling national databases
over the past several years to address
the issue. Further, unlike many previous
studies compiled solely by appraisers or
economists, we have leveraged expertise
in statistics, real estate valuation, and
economics to collaboratively develop
analytical models for those databases
that yield meaningful and reliable
results from the perspective of all three
areas of study. Combining significantly
larger data sets with more robust
modeling allows for considerably
improved study results.
Case Study to Address Impact
To illustrate the type of analysis that
refocused assignments can provide,
we considered proximity impacts
of electrical transmission corridors
on single-family homes in Salt Lake
County, Utah, where a significant
amount of research has already been
conducted on proximity to power lines.
Using a larger data set and more robust
analyses, we are able to address several
important issues that have been
inadequately addressed in previous
work.
While previous research focused
on specific neighborhoods or
subdivisions with certain data
limitations, our data covers almost
all single-family home transaction
sales in Salt Lake County from 2001
through 2014. Our transmission line
data contains location information for
each type of high-voltage overhead
transmission line, as well as medium-voltage transmission lines and
substations. In contrast to previous
studies, we are able to examine the
effects of one type of line, while
controlling for effects of others.
Due to limited data, previous
research had only considered a single
type of line or the combined effects
of various power lines. This led to
inadequately supported proximity
impact estimates. For example, if a
46kV corridor runs within 200 meters
of a 345kV transmission corridor,
ignoring the 46kV line may result in
proximity impacts relating to that line
being erroneously attributed to the
345kV line.
With more than 100,000 sales from
one county, we were able to consider a
wider range of property characteristics
than past studies. Also, previous studies
have substituted the influence of
macroeconomic conditions by simply
introducing time variables. In contrast,
we directly tested the influence of
macroeconomic conditions on the
influence of power lines on property
values. Additionally, our data begins
before the 2007 housing crash and
stretches through the recovery period,
thereby allowing for impacts to be
tested for market condition sensitivity
as well.
We also went beyond a typical
appraisal and applied tests for spatial
autocorrelation to investigate the
implications of Waldo Tobler’s
first law of geography that states,
“Everything is related to everything
else, but near things are more related
than distant things.” This analytical
step is crucial for improving the
extrapolative reliability of the study by
accounting for atypical neighborhood
or community market responses to
transmission corridors.
For Salt Lake County, we obtained
transmission line locations at regular
intervals for all individual transmission
lines by type as well as the location
of all substations. Using this data,
we calculated the distance from each
transmission line and substation
location data point to the location of
each of the county’s approximately
350,000 properties. This resulted
in more than 60 billion proximity
computations. Next, we calculated the
minimum distance from each property
to each transmission line.
We then matched each parcel to
the sales in our transaction database
during the study period. This resulted
in a combined database of all sales
information, including detailed
property characteristics, and distances
from each property to each type of
transmission line and to the closest
substation. We then combined the
resulting database with our economic
variable database to allow us to account
for market changes. In total, our model Adding recreational space can work to enhance the area under a transmission corridor.
