Affordable Housing Institute : US

From the Washington Post comes a summary of an important Brookings Institution study with a lurid title, Where Did They Go?, highlighting a major if perplexing demographic trend:

 

INDIANAPOLIS — Middle-class neighborhoods, long regarded as incubators for the American dream, are losing ground in cities across the country, shrinking at more than twice the rate of the middle class itself.

In their place, poor and rich neighborhoods are both on the rise, as cities and suburbs have become increasingly segregated by income, according to a Brookings Institution study released Thursday.

 

Brookings leaves no doubt that its researchers think this trend is bad: the report’s subtitle is The Decline of Middle-Income Neighborhoods in Metropolitan America.

It found that as a share of all urban and suburban neighborhoods, middle-income neighborhoods in the nation’s 100 largest metro areas have declined from 58% in 1970 to 41% in 2000.

 

In interpreting the information, we have to be very careful about whether our verbal descriptions still accurately describe the conditions observed.

 

“You keep on using that word … I do not think it means what you think it means.”

To that end, here are three groups of cities, chosen from the Brookings data. What organizing principle can you find for each of the three groups?

 

 

Answer later.

Indeed, I recommend you download and read the whole thing (available here in .pdf) because the actual data themselves paint a much more complex picture, one that I suspect is quite a bit different from the typical headline grabbed above.

 

Makes sense to me.

Middle-income neighborhoods — where families earn 80% to 120% of the local median income — have plunged by more than 20% as a share of all neighborhoods in Baltimore, Chicago, Los Angeles and Philadelphia. They are down 10% in the Washington area.

This invites the question, what is a middle-income neighborhood?

 

One possibility is that with wider income distribution (rich people’s average income is a higher multiple of poor people’s), our definition of a ‘middle-class neighborhood’ needs to changed.

 

Widening income inequality in the United States has been well documented in recent years, but the Brookings analysis of census data uncovered a much more accelerated decline in communities that house the middle class.

 

“No city in America has gotten more integrated by income in the last 30 years,” said Alan Berube, an urban demographer at Brookings who worked on the report.

[Brookings uses a precise definition of ‘city’. And ‘integrated’ is the wrong word; I believe Mr. Berube means ‘concentrated’ or ‘homogenized.’ I think the misnomer significant because it bespeaks a misunderstanding of what the data is saying.]

Before we can wring our hands about the vanishing middle-income neighborhood, we have to ask ourselves, even more fundamentally, what is a neighborhood?

 

Won’t you be my census tract?

Brookings defines it as a census tract, which the census defines as:

Census tracts generally have between 1,500 and 8,000 people, with an optimum size of 4,000 people. (Counties with fewer people have a single census tract.) When first delineated, census tracts are designed to be homogeneous with respect to population characteristics, economic status, and living conditions.

 

[Brookings also quite properly knocked out of its study very small tracts under 500 people, those with a heavy concentration of ‘group quarters’ (prisons, college dorms, nursing homes) as these have unrepresentative and deliberately homogenous populations.]

 

When laying its socioeconomic grid on America, the census itself started by creating economically homogenous subgroupings within communities. (The census provides complete lists here; click here to find your tract.)

 

The spatial size of census tracts varies widely depending on the density of settlement. Census tract boundaries are delineated with the intention of being maintained over many decades so that statistical comparisons can be made from decennial census to decennial census.

 

To convert the statistics into something comprehensible, I took a look at Our Fair City, my home town of Cambridge, Massachusetts.

 

Shaped like a butterfly, with Harvard Square the insectoidal head.

The City of Cambridge, with a land area of 7.1 square miles, is home to just over 100,000 people, divided into precisely thirty census tracts, an average of about 3,400 people per tract, each of which is about one-quarter of a square mile in area.

 

For example, here is my census tract; to gain a sense of its scale, find it on the larger map (under the word Wendell Street).

 

(Things you can learn on the web; in my census tract (3536), I am one of the 486 adults aged 45 to 54.)

Brookings therefore defines ‘David’s neighborhood’ as the 4,742 people who lived in a one-quarter-square mile part of Cambridge in the year 2000. Some of Cambridge’s thirty census tracts, I know from personal observation,

 

“You can observe a lot just by watching.”

are quite affluent. Others are much less so. This makes Cambridge, by Brookings’ standards, a very granular city — wealth and poverty (their terms, not mine!) rubbing against one another in distances of half a mile or so.

 

(To say nothing of deepest darkest Somerville, which as the map of my census tract reveals is breathing heavily down upon Cambridge from across the glass barrier of Beacon Street; see the map.)

 

However … isn’t the ideal city one that has all income levels?

[Continued tomorrow in Part 2.]

Send post as PDF to