Years ago Aaron Renn coined an urban paradigm that he labeled “the new donut.” The old donut, of course, is one that many of you will know well: poor downtown (hole in the donut) and wealthy suburbs (ring around the hole in the donut). This is a well documented phenomenon in many American cities.

The new donut reflects today’s return to city centers. It is the filling in – albeit only partially – of the middle of the donut. The reason I say only partially is because the data clearly suggests that, in many cases, there’s now a trough between the immediate core and the outer suburbs. 

In 2015, the University of Virginia published a study called The Changing Shape of American Cities. It looked at things like educational attainment and per capita income in 1990 and then compared it to more recent 2012-2015 data. But most significantly, it plotted this data against “miles from city center.” (I discovered this study via City Observatory.)

Here are educational attainment and per capita income for the 50 largest metro areas in the US. The orange line is 1990 data. The brown line is 2012 data. And the blue line is 2015 data. The x-axis is “miles from city center.”

Compared to 1990, it is clear that there has been noticeable spike in education and income in city centers. For the above composite index, more than 50% of adults over 25 now have a bachelor’s degree. But it has also accentuated the trough that appears to sit, on average, about 5 miles out from the center. 

In some metro areas, such as Charlotte (shown below), there has almost been a complete inversion. Education and income were highest 5 to 10 miles out from the center, but that has since flipped, along with a dramatic spike right in the center.

This is the new donut. If you’d like to see the graphs for all 66 American cities that form part of the study, you can do that here. 

Whenever I’m not sure what to write about, I just read. That’s one of the big benefits of daily blogging – it forces me to do that.

This morning I stumbled upon the blog of Jed Kolko. Jed is an economist and, up until 2015, he was Chief Economist and VP of Analytics at Trulia.

His most recent post argues – naturally with lots of data and charts – that for all of our talk of (re)urbanization, it’s actually a specific subset of the population that is far more likely to be have urbanized between 2000 and 2014: the young, rich, childless, and white. (Note: His post is talking specifically about U.S. cities.) 

Below are a few of his charts. 

In all cases, the x-axis represents % change in urban living between 2000 and 2014. All of the data is from Public Use Microdata Samples (PUMS) - 2000 decennial Census and from the 2014 one-year American Community Survey (ACS).

Here is age:

Household income:

Education and children:

And here is race/ethnicity:

In 1933, the United States Congress created the Home Owners’ Loan Corporation (HOLC). With foreclosures rising as a result of The Great Depression, the task of the agency was to provide new low-interest mortgages to both homeowners and private mortgage lenders. Between 1993 and 1936, the agency served about one million households.

By 1935, the parent company of the agency (the Federal Home Loan Bank Board) decided to initiate something called the “City Survey Program.” The idea was to look at local real estate trends – including the racial and ethnic composition of the country’s largest cities – in order to get a better understanding of how to manage all of these outstanding loans.

One outcome of this program was the creation of the HOLC’s infamous “residential security maps.” (Philadelphia’s is shown at the top of this post.)

These were maps that categorized city neighborhoods according to 4 grades. Grade A neighborhoods (green) were the best ones. They were ethnically homogenous and had room to be further developed. Grade B neighborhoods (blue) were the second-best ones. They were already completely developed, but were still considered desirable. Grade C neighborhoods (yellow) were starting to decline and showed an “infiltration of a lower grade population.” And finally, grade D neighborhoods were considered “hazardous” and colored in red. These neighborhoods had low homeownership rates, old crappy housing, and an “undesirable population”, which, at the time, largely referred to Jews and African Americans.

Some have argued that the HOLC and their “residential security maps” are what kicked off systematic mortgage discrimination in America’s inner city neighborhoods – later referred to as “redlining.” This was the practice of denying credit to people who lived in these undesirable neighborhoods (and even to real estate developers who wanted to build in these undesirable neighborhoods).

But University of Pennsylvania professor Amy Hillier has argued that these maps simply reflected the ethos of the time period. Using a sampling of HOLC mortgages, she found that 62% of them were issued to grade D (red) neighborhoods. The agency, itself, was not actually redlining in practice.

Furthermore, she also looked at private mortgages issued in Philadelphia between 1937 and 1950 and found that security grade rating actually had no impact on the total number of loans issued. She did, however, discover slightly higher interest rates for properties located near and in the bottom security grades.

All of this is to say that “redlining” is likely not the only culprit for inner city decay. There are other factors at play.

To that end, the National Bureau of Economic Research recently published a working paper, which I discovered through CityLab, called, “Racial Sorting and the Emergence of Segregation in American Cities.” The key finding here is as follows:

“Our preferred estimates suggest that white flight was responsible for 34 percent of the increase in segregation over the 1910s and 50 percent over the 1920s. Our analysis suggests that segregation would likely have arisen in American cities even without the presence of discriminatory institutions as a direct consequence of the widespread and decentralized relocation decisions of white urban residents.”

In other words, it wasn’t just mortgage discrimination; it was also just general discrimination. That actually makes a lot of sense, because, if you think about it, the former couldn’t have occurred without the latter being present.

Here’s how the research paper puts it (via CityLab):

“Policies that reduce barriers faced by blacks in the housing market may thus not prevent or reverse segregation as long as white households have the ability and desire to avoid black neighbors.”

(Note: Most of the information and data used in this post was sourced from the work and research of Amy Hillier.)