Over the past decade, gentrification has become a hot button in the politics of many American cities. Gentrification is the process of changing the character of a neighborhood through the influx of affluent residents and businesses. While increasing the economic vitality of these neighborhoods, it comes at the cost of displacement of the previous residents – especially those that were property renters.

The dominant pattern of urban areas for most of recorded history was consistent – rich people lived close to the action in the city center, and poor people lived at the edge of the city – understandable since the primary mode of transportation was walking. The industrial revolution left the cities noisy and dirty. By the turn of the twentieth century many cities had a growing number of “streetcar suburbs” and the widespread adoption of the automobile over the decades allowed even the middle class to flee to the new suburbs.

The construction of the Interstate Highway system had a profound effect on the cities as downtown areas were largely abandoned in the 1960’s and 1970’s in favor of ever distant suburbs. By the mid-1970’s, however, the signs of the resurgence of those downtowns were beginning to be noticeable, as the children of the suburbs began to return to the cities. Initially, this was welcomed by cash-starved inner cities, but as the trend continued – and accelerated over the past twenty years – these suburban refugees began to move into lower income areas, dislocating especially lower income renters.

The gentrification of a neighborhood comes with it both positives and negatives –

  • Reduction in crime
  • Reduced housing vacancy rates
  • Initially an increase in social diversity, followed by a decrease in diversity as the neighborhood ‘flips’
  • Increased property values and rents, which displaces the original residents, but results in higher municipal revenues

How do we identify areas which are in the process of gentrification? There is no single demographic indicator which identifies the phenomenon. We recently saw a map that used decreases in social security recipients to identify such areas, but the results are overly broad because there are many reasons why declines might be occurring that don’t qualify as gentrification.

To identify these areas, we used a complex set of variables to filter out the gentrifying areas more precisely by using variables on a relative basis. That is, we looked not at absolute changes in a variable, but the relative change in relation to the entire city. We settled on the following:

  • The median income had to be below average at the beginning of the study period (2010)
  • Relative increase in median household income
  • Relative decrease in vacancy rates
  • Relative increase in property values
  • An increase in income inequality relative to the city change, using the Snapshot inequality() function
  • A decrease in the housing stock built prior to 2010 relative to the change in households (that is, areas where the households have not declined, but the old housing stock has been at least partially replaced)

The results for Seattle are shown below, which parallel the precise neighborhoods identified in a recent Seattle Times article, “5 of the Seattle area’s most changed neighborhoods” (21 Jan 2020).

Three points should be made here:

  • Complex concepts are rarely measurable by a single variable
  • Care must always be taken with surrogate measures – if a variable isn’t directly measuring what you are trying to understand, you need to use it with due caution
  • Scale is important – a block group level analysis shows the high degree of concentration of the phenomenon, which tends to progress one street at a time and not over broad areas of the city simultaneously