Did
Here is the quick answer: The US Census Bureau reports that, for the 250 cities covered in the study,
Seattle ranks in the top 10% in terms of housing price increases from
1989-2006.
For those who are not convinced by the quick
answer, I provide a detailed discussion below using alternative housing price
data.
Before we look at alternative data, let me mention:
a)
The housing price data in the study was obtained
from the US Census Bureau. Census data is, in general, highly respected.
b)
The study discusses at length all different housing
price indicators and their advantages and disadvantages, and motivates the use
of the Census data.
c)
Whether housing prices increased a lot (or not) is
a different question than the one examined in the study, which asks “Did Regulations Contribute To Housing Prices
- And If So, How Much?” Think about it this way: even cities with small
increases in housing prices may find that a large share of this increase is
attributed to regulations.
Alternative
Housing Price Data: Standard &Poor’s Data
A popular, alternative housing price measure is the
Standard & Poor's
Case-Shiller Home Price Index®. The Standard & Poor’s
Methodology Factsheet highlights three important aspects of the
S&P data:
a)
It excludes all new construction, but it better
controls for housing characteristics than Census data (for example, whether the
size or quality of the average home has changed).
b)
It covers only 20 regions (vs. the 250 cities
covered by Census data).
c)
It does not represent cities, but metropolitan regions. S&P data for
What
does the S&P Housing Price Data Imply?
By default, the S&P data gives the year 2000 as
the “base year” (follow this link to the
MSExcel data). The data is easily plotted from 1990 – 2007 in
Figure1a, which is often displayed since it is a simple plot of the raw
data.
Figure 1a is, however, difficult to interpret
because all cities are plotted for a different base value at the data’s
starting point, in 1990. This may lead
to interpretations such as “after Seattle
started operating under growth management, housing prices increased less than
in other major metropolitan areas.”
Such statements are not quite accurate, however,
because Figure 1a shows the level of
housing prices; it is not a comparison of
growth rates of housing prices, which would be necessary to validate such
statements. Instead the reader must eyeball growth rates for all regions from
the housing price levels to judge whether Seattle’s housing price growth was
faster or slower than that of a given reference city or reference group.
The
issue is easily resolved by providing all cities with the same base year at the
starting point, which is the year 1990, i.e., let’s compare apples with apples.
Figure 1b clarifies that only one other region experienced a greater increase
in housing prices than Seattle since 1990: the Portland region. Except for the
Portland region, Seattle’s regional housing price increases have outpaced all
other major metropolitan areas housing prices since 1990.
The most effective (and simple) representation to
allow us to visualize and compare growth rates in housing prices is to plot the
growth rates explicitly. Figure 2 confirms that Seattle’s housing price growth
ranks second among major metropolitan areas since 1990. In Figure 2, I also
compare the growth rate of housing prices in Seattle for the S&P and Census
data, to highlight their similarity and consistency.
We
can go further and correlate the growth rate in housing prices with the Wharton
Land Use Index (plotted on the right-hand side axis) where higher values
indicate more restrictive land use environments. (Please keep in mind the
caveat that the Wharton data does not have full information on each region)
Figure
2
