University of Minnesota’s Edward G. Goetz, along with his colleagues Tony Damiano and Jason Hicks, argue that neighborhoods can – and should – be examined on both ends of the economic continuum. What’s more, they provide a metric for examining where economic outcomes and racial concentration intersect.
On one end are census tracts that they call “racially concentrated areas of poverty,” or RCAPs, that are more than 50 percent non-white and have poverty levels in excess of 40 percent. On the other are “racially concentrated areas of affluence,” or RCAAs, where at least 90 percent of residents are white and median incomes are at least four times the federal poverty level.
The authors applied these definitions in a study of 15 metropolitan areas. But do these definitions really make sense when looking at a place like Houston? I would say not necessarily.
What most caught my attention in this work was the extremely small number of RCAAs in Houston. Only 5 of the 1,069 census tracts in the greater Houston area – just 0.5 percent – are considered areas of racially concentrated affluence.
Those and related numbers suggest that Houston is among the metropolitan areas with the lowest representations of racially concentrated affluence (see Figure 1). Similarly low representations of RCAAs were also found for Los Angeles, Miami, and San Francisco.
What’s curious about this result is that Houston – as well as Los Angeles and Miami – are not without high concentrations of wealth. In fact, based on a recent Pew report, Houston has the highest concentration of wealth among large cities.
So maybe Houston’s concentration of wealth is just disconnected from concentrations of whites.
But maybe not. There’s an important distinguishing feature of Houston and the other low-RCAA cities that needs to be considered – the metropolitan areas with a low representation of RCAAs are also some of the most racially/ethnically diverse areas in the country.
Based on 2010 population data, the average percentage of the total population that is non-Hispanic white is only 37.1 percent in the low-RCAA areas. On the other hand, the metropolitan areas with the highest percentage of the population living in RCAAs – like Boston, St. Louis, Minneapolis and Philadelphia – are 72.4 percent non-Hispanic white, on average.
This is a drastic difference in the percentage of non-Hispanic whites – a 35 percentage-points difference. And it could have major consequences for how well a 90 percent threshold can work when defining RCAAs. White residential areas that include other groups are, by definition, more likely in cities where those groups comprise a larger share of the total population.
This potential methodological issue is made even clearer when we consider the presence of RCAPs in Houston.
In contrast to the small number of RCAAs, Greater Houston has 63 tracts that are considered RCAPs. So whatever explains why we have low racially concentrated affluence is not similarly affecting racially concentrated poverty. And a glaring difference between RCAPs and RCAAs lies in how we are defining racial concentration – 50 and 90 percent, respectively.
These concerns lead me to question whether we can answer questions about racially concentrated affluence in diverse cities with the proposed 90 percent threshold.
I suggest that what is considered a “high” or “disproportionate” white concentration should be tied to the composition of the city as a whole. What this would entail is separate thresholds for cities with comparable white concentrations. Or perhaps more appropriately, metropolitan areas could be grouped by levels of racial/ethnic diversity.
Starting simply by using the percent white values, let’s assume that the 90 percent threshold makes sense for the high-RCAA cities – Boston, St. Louis, Minneapolis and Philadelphia. Based on their average white population size of roughly 70 percent, there is a 20 percentage-point difference between the metropolitan average and the 90 percent threshold.
Even if we used this basic approach, then a similar difference would suggest that we use a 60 percent threshold for the more diverse cities.
That’s a full 30 percentage-points lower than the proposed threshold. And there’s a case to be made for more complex approaches that would suggest an even lower threshold.
So before rushing to embrace new measures for studying racially concentrated affluence, we should take a step back and consider what they mean in the context of diverse cities. And as the demographics of Houston and similar cities show, our approach may not be the same for all places.