Changes in racial composition

Antiplanner reports,

Between 2015 and 2016, the total population of the San Francisco-Oakland urban area grew by 13,773 people, but the black population shrank by 5,839, suggesting that Bay Area land-use policies continue to push low-income people out of the region by making housing unaffordable. The Austin urban area, to its shame, saw a decline of 4,439 blacks despite a total population growth of 25,316.

Race is a complicated issue, made more complicated by the increasing (and healthy) mixture of races. According to the 2016 American Community Survey, the number of Americans who are “white alone” declined by 296,061 in 2016, while the number who identify themselves as “two or more races” grew by 445,000; some of the decline of the former and growth of the latter is probably because people are more willing to self-identify as being of mixed races.

In the past, I’ve used blacks as a bellwether of housing affordability problems because black per capita incomes have consistently been about 60 percent of whites’. I’ve previously used “black alone,” but this year that produced some odd results: both white alone and black alone populations declined in sixteen different states. For example, California’s total population grew by 105,000, but its white-alone population shrank by 404,000 while its black-alone population shrank by nearly 12,000. It seems likely that most of the changes in white-alone and black-alone numbers are due to redefinitions, not migrations.

Basic numbers for seven different races–white alone, black alone, Native American alone, Asian alone, Native Hawaiian alone, other races, and two or more races–can be found in American Community Survey table B02001. I’ve downloaded this table for 2006, 2010, 2015, and 2016 for the nation, states, counties, cities, and urban areas and copied the population numbers for into a single spreadsheet. A summary sheet has data for total, white alone, and black alone.

American Community Survey table B01003 shows the total population while B02009 shows the population of “black alone or in combination with some other race.” So I downloaded these tables for the same years and geographic areas and combined them into one spreadsheet for the total population and one for black populations. Even then, I am not certain of the results. I’m a little surprised, for example, that the total population of the Los Angeles urban area declined by 23,106 people while the black population grew by 22,214; I would have guessed it would be the other way around.

Some of the changes in total state populations suggest that the 2016 survey may not be reliable. According to the survey, 13 states lost populations, with Illinois leading at decline of 58,456 people, followed by New York with 50,502 fewer than in 2015. These numbers don’t agree with the official population estimates, which say that only eight states lost population. Illinois still leads at minus 37,508, but New York only lost 1,894. The differences are almost entirely due to adjustments to the 2015 population numbers. Until the 2016 numbers are adjusted, it is probably unsafe to use them as indicators of changes in racial compositions.