What the Maps Tell Us: Initial Findings

The first two maps in this series provide information about owner occupied housing: Median Home Value and Change in Median Home Value (2000-2010). The census tracts with the highest median housing prices are in the west hills, Lake Oswego, Irvington and Hayden Island. The median sales prices in 2010-11 ranged from approximately $535,500 to $806,000 in these areas. The census tracts with the lowest median housing prices include portions of north Portland, outer east Portland,  east Multnomah County, and portions of Oregon City, Canby, Tualatin, Beaverton, Hillsboro, Forest Grove and Vancouver, as well as the city of Cornelius. Median sales prices in 2010-11 ranged from approximately $140,000 to $257,000 in these areas.

Nearly all the census tracts where the median housing sales price more than doubled from 2000 to 2010 are located in the cities of Portland and Vancouver. These areas include the Vancouver neighborhoods of Esther Short, Hough and Arnada. In Portland, these areas include neighborhoods in inner northeast and southeast, the southwest neighborhoods of Markham, Arnold Creek, Marshall Park and Southwest Hills, and the northwest neighborhoods of Forest Park and Linnton.

Census tracts where the median inflation-adjust housing sales price was lower in 2010 than in 2000 include a few neighborhoods in outer east Portland and east Multnomah County and a census tract that includes portions of Forest Grove and adjacent unincorporated Washington County. 

These data do not fully capture the volatility of housing prices over the past decade. Prices rose rapidly in the early part of the decade and then dropped toward the latter part of the decade. Prices are now beginning to gradually increase again. The Change in Median Home Value map compares sales price data on homes sold in the year 2000 and 2010, so it reflects two snapshots in time rather than the trends over the course of the entire decade.

The Median Rental Cost map provides data on rental housing; unfortunately these data are only available by zip code, so the map does not provide as nuanced an analysis as the other maps in this series. It is also important to note that the data used for this map are taken from a relatively small sample of units, and the units included may or may not be representative of the full range of rental housing in the region.

The zip codes with the highest rents in the region include Lake Oswego, downtown Portland and the adjacent Southwest Hills, Pearl and Old Town neighborhoods. Median monthly rents ranged from $1,150 to $1,580 in these areas in 2011 for a two-bedroom unit. The inclusion of Lake Oswego, the Southwest Hills, and the Pearl District among the region’s highest rent neighborhoods is not too surprising. The inclusion of downtown Portland and Old Town is more surprising, since the region’s highest poverty census tract overlaps with both of these neighborhoods.

The zip codes with the lowest rents include portions of Estacada, unincorporated Clackamas County, Cornelius, unincorporated Washington County, Vancouver, Battleground, and unincorporated Clark County. Median monthly rents ranged from $479 to $650 for a two-bedroom unit in these areas in 2011.

The Housing Purchasing Power map indicates whether or not the average household living in a particular census tract would be able to afford an average home for sale in that census tract at current day sales prices. In Clark County, the least affordable census tracts (the places where current residents are least likely to be able to afford to purchase an existing home) include an area comprised of the Fruit Valley and West Hazel Dell neighborhoods and the unincorporated area immediately to the north, and census tracts along State Road 500 between I-5 and I-205. Portland’s least affordable census tracts include those in the St. Johns neighborhood in north Portland and the Lents, Brentwood/Darlington, Southgate, Powellhurst/Gilbert, Montavilla, Centennial and Sumner neighborhoods in outer east Portland. Other least affordable census tracts around the region include those in Central Beaverton and the North Gresham, Rockwood and Northeast neighborhoods in Gresham.

There are several potential reasons why an area would not be affordable to current residents. This divergence might be explained by housing values rising more rapidly than the residents’ incomes (raising potential future gentrification concerns), by the fact that a large share of households are renters who are unable to afford to purchase homes, by an influx of lower-income households into a historically moderate income neighborhood (raising potential concerns of disinvestment), or by other factors. Looking at this map in conjunction with some of the other Atlas maps may provide additional insights into the mismatch between housing values and the purchasing power of existing residents in these areas. 

For many years, housing has been considered to be affordable if it costs no more than 28% to 35% of household income. More recently, analysts have begun to consider the affordability of housing and transportation costs jointly because housing and transportation are usually a household’s two largest expenses, and households often make tradeoffs between the two. For example, a household may choose to live in a more expensive apartment because it is more convenient to destinations that they frequent regularly, such as work. Other households may choose to live in a less expensive housing location but pay more for transportation costs because the home is farther away from frequent destinations. 

The map Housing and Transportation Cost Burden utilizes a complex modeling algorithm to determine the combined housing and transportation costs relative to household income in each part of the region. The data are mapped by Metro Housing Needs Analysis Subareas, which are aggregations of census tracts. The map shows what residents in each subarea pay for housing and transportation relative to the median income for that area. In this analysis, a household is considered cost-burdened if it spends more than 45% to 55% of its income on housing and transportation costs (45% for a low income household).   

The map shows that the highest combined housing and transportation cost burden in the region is found in an area that includes Forest Grove, Cornelius and the surrounding unincorporated area of west Washington County. The combined cost for housing and transportation is 52.2% of that area’s median income, which is considered unaffordable. The area with the least cost burdened residents is located around Happy Valley, where combined housing and transportation costs represent 37.4% of that area’s median income. However, given the diversity of incomes and housing costs in that area, this average figure may mask the experiences of the area’s lowest income residents who may bear a much higher combined housing and transportation cost burden than their wealthier neighbors.

The map Location of Publicly Subsidized Affordable Housing in Relationship to Areas with Above Regional Average Percent Populations in Poverty depicts the locations of housing units that have received public funding in order to maintain rents that are affordable to households that cannot afford market rents. The map shows that nearly all of the region’s publicly subsidized housing units are located in areas where the poverty rate is above the regional average of 6.72%. This suggests a general alignment between the locations of affordable housing units and the people who are in need of that housing. But it also suggests that many affordable housing units may be located in areas of lower opportunity, since high poverty neighborhoods frequently have reduced access to opportunity. It is important to note that this map does not include information about where people live who use subsidized housing vouchers to help pay for market rate housing units.

Housing affordability plays a central role in equity issues related to access to opportunity. Where people live has an effect on their day-do-day lives, their access to jobs, quality education, and a healthy environment, and their social networks. Housing patterns that separate low income households or people of color from access to opportunity pose major equity issues. Housing affordability—the share of income that people pay for household costs—also affects the degree to which residents have sufficient resources to take care of life’s other necessities.  The maps in this series help set the stage for further equity discussions of the region’s housing, transportation and land use policies.