The Urbanist Musings of Pete Saunders

CSY At Forbes — Where Gentrification’s Impact Is Greatest, Part II

Illustration of a proposed residential tower in Jersey City, NJ.  Source:

(Note: This is part two of a short series that initially ran at my Forbes site and is being republished here.  Followers of mine will recognize it covers a familiar theme; you can’t apply a one-size-fits-all policy approach to cities when their experiences vary so much.  Please take a look.  -Pete)

It’s well known that not all cities nationwide share in the success experienced by those on the coasts. While some urbanists wish to alter the way development occurs in their cities, others are still seeking ways to stimulate it. But do we really know where those differences exist? Is there a way to compare and contrast how gentrification is impacting a wide range of cities?

My analysis of Census data to determine the U.S. cities where gentrification has had the greatest impact — determined in this case as city gains in median household incomes relative to their overall urban areas — continues. My analysis of the 14 largest U.S. urban areas (those with a population over three million as of 2010), can be found here.

A quick recap: I started an analysis with a couple of simple premises. The first premise is that a simple definition of gentrification is that it represents the transition of a low-income neighborhood into a high-income neighborhood. The second premise is that, within metro areas, household incomes tend to be lower in central cities and higher in suburban areas. For analytical purposes, the thinking is that if we examine changes in household income over time, at both the city and metro area levels, we should be able to get an idea of cities where the influx of new residents and new money has had the greatest impact — where city household incomes are closing the gap with their respective metro areas.

My analysis started by looking at U.S. Census Bureau American Community Survey data between 2000 and 2015, looking at median household income figures for the 14 largest urbanized areas in America, and their respective central cities (that includes all U.S. urbanized areas with more than three million people). I now have data on the 27 additional urbanized areas with populations above one million people. My first cut at this looked at household income data in 2000, 2010 and 2015; I’ve since added 2005 data to equalize the intervals and get a truer sense of change both before and after the late 2000s Great Recession.

Why the focus on urbanized areas? They’re dynamic: they are comprised of census tracts of contiguous developed areas centered on cities with a population density of 1,000 people per square mile or more. They’re also more adaptable to comparison than the broader metro area definition, which pulls whole counties into the mix and includes parts that might be best described as exurban or even rural. Taken together, the 41 largest urbanized areas with more than one million people contain 43 percent of the U.S. population. By themselves, the 14 urbanized areas with more than three million residents contain more than 28 percent of the nation’s population.

My findings for the 14 largest urbanized areas confirmed that large cities are closing the city/suburban income gap. Of the 14, ten posted gains in city median household incomes, with four — Washington, DC, Atlanta, Seattle and San Francisco — showing substantial gains relative to their overall urbanized areas. In 2000, the top 14 urbanized areas had cities where median household incomes averaged 76 cents for every dollar earned within the entire urbanized area, with no core city having a median income than its urbanized area. By 2015, cities were earning 80 cents for every dollar earned within their urbanized area, and two cities, San Francisco and Seattle, had higher median household incomes than their urbanized areas. The cities that missed out on this transformation, Dallas, Houston, Detroit and Phoenix, likely missed because outward expansion continues to rule their development patterns (Dallas, Houston and Phoenix), or because of structural economic change (Detroit, although a case could be made for the former here, too).

Mid-sized urbanized areas, however, with populations between one and three million, are experiencing a different reality. What I found:

  • The central cities of the 27 urbanized areas with populations between one and three million closed the median household income gap by just 0.2% from 2000-2015. In 2000, the core cities of this group had a median household income that averaged 81.9% of the total urbanized area median household income ($37,339 to $45,463). By 2015 the gap between them and the suburbs barely moved, averaging 82.2% of total urbanized area median household income ($49,378 to $59,974).
  • Only 13 of 27 core cities in this group gained income relative to their suburbs. The top five gainers — St. Louis, Portland, Minneapolis/St. Paul, Riverside/San Bernardino and Denver — closed the city/suburban median household income gap by an average of 9.5% between 2000-2015. However, the bottom five laggards — Milwaukee, San Jose, Cincinnati, Cleveland and Indianapolis — saw their city incomes falling 8% behind suburban ones over the same period.
  • When it comes to gentrification, size (and location) matters. Once again, the data confirms what many people intuitively know about major cities already. Of the largest 20 urbanized areas, 14 core cities made income gains relative to their suburbs. However, only nine of the next 21 largest urban areas showed similar gains. Related, 13 of 17 urban areas in the Census-designated East and West regions of the nation made income gains, while only 10 of 24 Midwestern and Southern cities made similar gains. And that includes two urban areas that are now only nominally southern, Baltimore and Washington, DC, and are now more economically and culturally linked with other cities in the northeast corridor than they are with Southern cities.

And of course, there are tables to tell the story. First, the 14 largest urbanized areas:

And then, the next 27 urbanized areas:

And lastly, the super table that includes all 41 urbanized areas:

There are, of course, limits to using this data for making grand findings. Not all cities are created equal; some consist largely of pre-WWII development that is dense and walkable, while others are mostly post-WWII development that is auto-oriented and has more in common with their suburban counterparts. Nor are all urbanized areas equal; some have core cities that comprise as much as half of the urbanized area’s population, while others have small core cities and a vast suburban landscape. As for interpreting the data itself, it’s clear that all changes can’t be directly attributed to gentrification, or a lack of it. City incomes could rise relative to suburban areas due to increases in already high income areas, meaning rich areas are becoming richer. And city incomes could still fall relative to suburban areas even when there is a significant influx of wealthier residents, if the middle class continues to flee in even larger numbers.

But there’s a simple reason this data is useful, even important. There are many urban policymakers who have focused their attention on the challenges of coastal cities, who in some respects have become victims of their own success. However, it’s apparent that many cities in the middle of the nation have yet to experience the same success. Policy recommendations geared toward coastal city challenges may have little to no impact in our interior cities. In fact, they may need a set of policy recommendations all their own.

2 Responses to “CSY At Forbes — Where Gentrification’s Impact Is Greatest, Part II”

  1. Anonymous

    Median household income (MHI) as a measure has a natural bias against core cities, at least cities that have lots of young adults. The mere existence of studio or 1-bedroom housing stock will reduce an area's MHI even when average wages per adult are identical, since small apartments are unlikely to house multiple wage earners.For San Francisco's MHI to exceed the urbanized area's, the average capita income in SF probably vastly exceeds the income in the urbanized area.Not that this changes the results of the analysis, but recognize that it is not an even playing field.


  2. FdoCenteno

    When you use \”gentrification\” you cite residential dynamics, but there's also commercial gentrification. How does that play out?When you say \”city\” you really mean regional (i.e. \”urbanized\”), using metro data. My problem with \”metro\”, as where I live, this area involves 8 surrounding (mostly rural) counties, encompassing 7,400 sq. mi. I'm much more interested in my actual \”city\” metrics & dynamics than \”metro\” data.Our city planners & officials also love to talk about regionalism, which waters down what is actually happening in the city, and ironically, \”city\” taxpayers are paying dearly for this heavily subsidized expansion, where success is measured in business terms, rather than in socioeconomic terms. Enough with that. This is the conceptual difference between \”urban\” planners & community economic development planners; the first values the built environment, the second values increasing the standards of living & quality of life outcomes, in real terms.Which side on you on?



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Basic HTML is allowed. Your email address will not be published.

Subscribe to this comment feed via RSS

%d bloggers like this: