Observations on Local Development Cycles

9-2-99

Dear Clients & Friends,

What follows are my observations on a single aspect of land development patterns in our area: the variety of age in urban housing stock compared to the inventory in suburban neighborhoods.

One market segment that historically enjoys among the highest demand and lowest supply characteristics is the urban, single family housing market in Northeast Seattle. Part of this appeal is likely attributable to its close proximity to downtown Seattle. But, there are locations that have closer proximity to the downtown core. The architectural design and construction quality in Northeast Seattle is sound, but many other areas have better examples of period design or better finish quality. A number of locations around Northeast Seattle have marketable views, but plenty of others around the region have far more impressive view amenities.

After more than twenty years of studying the Seattle housing market, it is my opinion that the health of the urban residential real estate market is positively influenced by the diversity of age in the housing stock. Seattle is a relatively new city as American cities go, with most of the existing housing built after 1900.

Suburban development came about 40 years behind construction of the bulk of the city’s existing housing stock and, like other areas around the country, occurred with more concentrated patterns. That is, many houses were built in a short period of time in a small geographic area.

The first table below, "Development Pattern A," shows a typical suburban plat development cycle. Hypothetically, we begin in the 1960’s when a developer buys 40 acres of raw land from a local farmer. During the five-year period of initial development and construction, 120 single-family houses are sold. This is represented by the table’s first rise in the chart line.

When construction is completed, the neighborhood’s development trend evolves into its period of stability, which, with good construction practices, lasts between 20 to 25 years. At the end of this period of stability there are 120 houses with major components such as a roof, furnace, kitchen, baths, etc. that are likely in need of repair or replacement. Please see the previous letter discussing long-term market cycles. Since at least the 1960’s, if not earlier, the Central Puget Sound area has experienced a market peak every ten (+/-) years, which is followed by a sudden drop in selling activity and price corrections.

If, at the end of the period of stability discussed above, a homeowner attempts to sell, and the selling period coincides with a low point in the region’s long-term housing market cycle, these combined influences will likely exacerbate the downward pressure on prices. Even if the market is nearing the peak of volatility and the house brought to market has been updated, declining physical characteristics of the surrounding properties (regression, in appraisal terminology) will still have a negative influence on price.

In an alternative scenario, and one to which I attribute at least some of the long lasting health of Seattle's urban residential neighborhoods, is overlapping development patterns. This practice is depicted in the table labeled "Development Pattern B."

If construction of the housing inventory is undertaken on more of a "scatter site" basis, i.e. some houses in a location are built in 1960's, a number more in the next decade, still more the following decade, and so forth, there is less risk of wide spread neighborhood deterioration. Sixty or more years later, re-generation of the housing stock can successfully occur on a scatter site basis as the land value has increased to the point where the lower quality houses no longer contribute sufficiently to the total value and are replaced by new construction.

The reality of the suburban development business makes this second kind of pattern very difficult to achieve. Regulatory restrictions, financing constraints, cost management, and marketing of planned communities, have all influenced the plat development patterns evident across the country since the 1950's.

The table below details a comparison of urban and suburban neighborhoods and further illustrates the concepts discussed above. Using Transamerica Intellitech's Metroscan database as our information source, we compare a portion of the Ravenna/Bryant area of Northeast Seattle to the Kirkland/Redmond area just to the north and east of downtown Bellevue.

These areas are similar in that they are mid to upper priced locations situated adjacent to more expensive housing areas closer in to the retail and employment centers of downtown Seattle and downtown Bellevue, respectively. The median prices of the two subject neighborhoods are both approximately 55% of their respective more expensive, closer in areas. The samples were similar in size, totaling approximately 1,900 houses in the suburbs and 1,800 in the urban setting.

Note that in suburban housing stock, 90% of the construction took place over three decades while in the urban sample 90% is spread out over five decades of building activity.

Disclaimer: Please note, that this is a commentary on development patterns and their long-term impact in the marketplace. This commentary is not intended in any way to disparage one area or glorify another. On the contrary, both of the data samples submitted do, in fact, represent marketable and desirable residential locations in our region. For the future, the alternative perspective, Development Pattern "B" above, may be worth discussing at zoning, land use and community planning meetings.

Regards,

Robert E. Rothstein