Clark County, Washington is situated immediately to the north of Portland, Oregon across the Columbia River and is considered part of the Portland-Vancouver consolidated metropolitan area. Vancouver (Clark County) is considered a PMSA in its own right. However, Vancouver and the rest of Clark County have served as urban growth outlets for Portland since at least 1979, when Portland established urban growth areas as part of Oregon's statewide growth management law of 1973. In 1994 Vancouver and other incorporated areas of Clark County established final urban growth areas as mandated by Washington's Growth Management Act (GMA) of 1990, effectively closing urban area expansion in the entire metropolitan Portland region.

Comprehensive planning guidelines under Washington's GMA require that urban growth areas (UGAs) need to accommodate twenty years of growth. Land located outside of UGAs is reserved for "non­urban" use, including agricultural, forestry and recreational activities. While land contained within the UGAs needs to accommodate anticipated growth, future growth can be expected at greater densities and location choice will be limited within the UGAs. Increased densities and limited choice of location for residential uses within UGAs may cause land price inflation due to a perception of shortages in developable land. Also, since nearly all growth will be restricted to UGAs, demand for exurban residential development will be met with no new supply. With the exception of lots platted prior to implementation of UGAs, exurban residential development will be restricted to 5 acre minimum lots, and be subject to natural resource, agricultural and recreational set asides.

Previous research demonstrates that once urban growth controls are applied uniformly across a jurisdiction, residential lot and house prices experience significant inflation. Subsequent theory leads to two related research hypotheses addressed in this study. First, there will be a significant and positive residential lot price effect resulting from implementation of the Growth Manage-ment Act in Clark County, Washington. Second, the significant positive price effect will occur both inside and outside the UGA.

Two of the anticipated effects are substantiated in the current study, with an overall lot price increase of 35.5% after implementation of urban growth areas. Lot prices increased slightly higher within the UGA (38.7%) compared to all lots examined. A significant outside/post UGA lot price increase was not substantiated by the data. In this final case prices seem higher, but a limited number of observations limited the statistical robustness of the model.

To isolate the impacts of growth management boundaries, a theoretical model was developed which included economic factors influencing the greater Portland market generally and Clark County specifically, along with characteristics of the lots themselves. The economic factors included population, interest rates, wage rate, a proxy for construction costs (CPI) and unemployment trend. Physical characteristics of the lots included in the model were size in acres and distance from the Portland central business district (CBD). Other control variables were quarter of the year in addition to the inside/outside UGA identifier. The unemployment trend variable entered the model for the current period. Other economic terms were lagged one quarter. The dependent variable was the log of the sales price per acre.

The model presented analysis problems which are frequently encountered in economics, primarily the high degree of correlation between the economic terms. Accordingly, subset models were estimated to isolate the impact of each economic factor. All economic terms were separately significant, with the anticipated sign. Finally, three analysis of covariance models were estimated. The first dealt solely with post-UGA effects, disregarding the placement of the lot inside or outside the UGA. The second model dealt only with inside UGA properties, while the final model dealt only with outside UGA lots. For the complete sample and the lots located inside the UGA there was a statistically significant increase in prices due to the implementation of GMA. In addition, lots sold in the second half of each year commanded higher prices than those sold in the first half. Per acre prices were lower for larger lots, but the relationship between price per acre and lot size was nonlinear, as demonstrated by the significant and positive sign on the lot-size-squared term. For the lots located outside the UGA, there appeared to be higher prices after the GMA implementation, but the result was not statistically significant perhaps due to the small outside the UGA sample size.

One implication of this study is that implementation of the GMA, particularly the component of GMA which establishes urban growth areas as part of local comprehensive plans, may be incompatible with one goal of the law, "[to] encourage the availability of affordable housing to all economic segments of the population . . ." Based on an average lot price of $43,282 prior to establishment of the final UGA in Clark County, the county­wide increase of 35.5% in price after UGA implementation translated into a $15,365 increase in the price of a typical lot. This increase in price is sufficient to deny access to new housing to many consumers. In addition, as lot prices increase builders often feel compelled to build more costly homes on the lots to keep the land component of total housing cost within normal ranges. Further, as higher lot prices impact the overall local housing economy, the price of existing homes may also increase.

This research began as an effort to compare Washington's new growth management policies to Oregon's established program. However, problems obtaining a satisfactory sample of Oregon properties, coupled with an evolving focus on the impact of closing the growth outlet shifted the focus to Clark County. By the time these problems were identified, neither time nor the financing for this project offered the opportunity to develop refined data sets from multiple sources. Further studies of lot prices should incorporate information from multiple listing services and should include other characteristics of the individual lots, such as view, develop-ment difficulties, road access, distance to schools and amenities, etc.

The results of this limited study also lead to the question of whether there was an immediate impact on the price of improved properties (homes). Future studies should include both Clark County and other Portland-area communities so see if there was an impact on overall housing affordability which can be attributed to closing the Clark County growth outlet by implementing Washington's Growth Management Act.