The price of “somewhere”: Gentrification and Equality

The title, as long time readers have already figured out, plays on my favorite geography book – James H. Kunstler’s The Geography of Nowhere: The Rise and Decline of America’s Man-Made Landscape.The book, written in the mid-/late-1990s, is an excellent perspective on the history of the United States’ built landscape, specifically the typical American suburb (of which Z Geography is intimately familiar). The book reads as a eulogy for the “classic” American small-towns of the northeast. The “Main Street” that modern politicians are so fond of recalling. These towns and villages were small, reasonably compact, intimate, and green. In other words, the complete antithesis of the modern suburb.

In this post, we’ll discuss how these “Nowheres” are in transition to becoming “Somewheres” in both cities and suburbs. In highlighting the disparate, but legitimate, interests associated with development as well as its pitfalls, I’m hoping to draw your attention to the importance of these political choices. Beyond politics, there is also room for decency.

As any urban resident will tell you, things are changing. White Americans are returning, following the well-discussed “White Flight” of the 1950s and 1960s, when the racial group left cities for safer, cleaner, newer suburbs. The process of returning and the development associated with it has been coined gentrification. Gentrification involves an increase in home values and better delivery of public goods (e.g. education, transportation, security) – city managers like the increased tax revenue. Young, white families like being in a city while enjoying more effective utilities. Of course, there are already residents living in these cities – black home owners and renters, with decidedly less income, who are eventually “encouraged” out by developers or high property values.

These disparate outcomes increased development while original residents are pushed out have been discussed in some media. Recently, the Washington Post described how large, predominantly black, churches in Washington, DC are attempting to prevent the creation of bike lanes on the streets near the church. As the Post observes:

  • “the packed meeting…highlighted a tension in the rapidly changing District [of Columbia] between longtime, black residents and new, largely white residents.”
  • “These conflicts stem from the change in D.C.’s neighborhoods. Many of D.C.’s churches were built at a time when their neighborhoods, such as Shaw, weren’t as teeming with condos and restaurants,  and parking wasn’t as big of an issue. Additionally, many longtime churchgoers have left the city for the suburbs and now commute to their old churches by car.”

The issue relates to a proposal to establish bike lanes on streets near the church, thus removing parking spaces. While not summarized as such the “conflict” (because that is what it is) is between two distinct social groups (blacks, whites) on how best to use (in this case transportation) a finite, terrestrial resource (land). Again the Post:

  • “This ain’t London, this ain’t Europe. The United States is built on the automobile and we need to respect that,” said Michael Green, a deacon at New Bethel Baptist Church.
  • “Washington Area Bicycle Association, a group that advocates for cycling in the city, argues that bike lanes wouldn’t prevent anyone from going to church.  There are other modes of transportation available to churchgoers, and bike lanes are necessary for the safety of the city’s increasing number of cyclists.”

The conflict, obviously, is ultimately a political question and as the Advisory Neighborhood Commission member notes in the article not everyone is going to be happy with the outcome. The Washington Post article presents both sides of the dispute as legitimate stakeholders with competing interests. Often, it is too easy to characterize (or spin, if you like) these conflicts into easily identifiable “right” and “wrong” positions under a completely different narrative, one for each stakeholder.

Gentrification is not only happening in America’s cities, it is also happening in some of the suburbs as well. Considering that many of suburbs are still more affluent (and house predominantly whites) than the cities they surround, what I’m really referring to is the development or transformation of these communities.

Politico describes the ongoing transformation in a suburb outside Chicago, Evanston, Illinois, which trying to “kill the car” as the title notes. The article is a 5-page look at a new-old concept: transit-oriented development (TOD). New because, Americans (sometime after Europeans) “rediscovered” it. Old because we employed this concept with railroads and streetcars. Evanston’s planning (since 1986) involves mixed-use development (i.e. residential and commercial) close to one another near a transit node (hence transit-oriented development). The theory is to promote walkability within insulated communities – walk from your apartment to the grocery store, bakery, or school. But, if your job is in Chicago, then there is a light rail just down the street. The net effect, of course, is to make cars useless.

Unfortunately, Politico gives decidedly short thrift to the problems of development and gentrification (you have to dig to page 5). According to the article, critics of TOD refer to it as “transit oriented displacement.” There is only a general mention of “affordable housing” as a potential explanation to how TOD, with its rising property values and prices, generates “diverse neighborhoods—diverse in population, retail, entertainment and housing”.

At the moment, according to the 2010 Census, Evanston was more racially and ethnically diverse compared to the rest of Illinois (66% White, 72% White respectively). However, the city is also markedly more affluent based on average income and with median home values of over $350,000 compared to $182,000. There is already a bar to live in this transforming suburb where a car is not as necessary. The unintended consequence, without guaranteed affordable housing, is to permit an elite core of residents the luxury of short commutes, walkability, and liveability, while a much larger group of workers must commute in – probably with cars – because they can’t afford it.

There are no easy solutions to these political problems. How do we as a local electorate balance the desires of newcomers with the desires of long-time residents? Many would argue that these questions should be decided at the ballot box – but that is also a tyranny of the majority. A time may not be far off when your majority becomes a minority – what then will we have built? What precedent would you have set?

Visualizing Gentrification: an economic Geography perspective

In my last post, I mapped and discussed gentrification in the District of Columbia from the point of view of race and ethnicity. I showed, to quickly summarize, the prevalence of the “White alone” population group in gentrifying neighborhoods in the mid-Northwest and Capitol Hill. Of course, relying purely on racial/ethnic data is a recipe for disaster and I questioned whether or not gentrification was occurring within older racial/ethnic communities. In this post, I present and discuss this economic geography perspective of gentrification in D.C.

As always, let’s start with the data. As you probably already know, I’m utilizing freely available data from the U.S. Census Bureau. Unlike last time, however, I’m operating at a different scale and with different products. Whereas I used the two Decennial Censuses and the block group in the race/ethnicity post, I’m forced to use the 2000 Census and the 2011 American Community Survey (ACS) for this one, at the Census Tract level. Apparently, Census didn’t include income questions on the 2010 Census (I wouldn’t know since I never got a Census form) since they had been a part of the ACS since 2008 (or so). In addition, information at the block group level was supposed to be available by 2010 but I searched American FactFinder (…) and couldn’t come up with anything related to income, at the block group level, except for the 2000 Census. So I moved to the Census Tract scale, still effective for community analysis and the data was available from the 2000 Census and 2011 ACS. Sometimes folks you can only analyze what the data gives you… At any rate, I pulled average Median Income across all households in a census tract in 1999 and 2011.

The first graphic below depicts median income by census tract in the District for 2000 and 2011. The images paint a fairly rosy picture overall, between 2000 and 2011 most areas of the District experienced growth in the median income. Particularly dramatic is the far western and northern areas of Northwest and Capitol, which experienced a number of census tracts reaching over $100,000 for median income. Even census tracts in the more economically depressed areas of the southeastern quadrant (particularly across the Anacostia River) experienced income growth moving many areas from below $25,000 a year to between $25,000 and $40,000. Most interesting from the point of gentrification, however, is the mid-Northwest where some census tracts with median incomes between $25 and $40 thousand a year in 1999/2000 were in the $60 to $99 (and higher) category by 2011. Give this some thought, over a little more than ten years the median household income (that is the income earned by the “middle” household when incomes are rank-ordered) increased from say $33 thousand a year to $80 thousand a year. For a bit of description, the average median income in D.C. in 2000 was about $42,000, equivalent to the “yellow category”, with a minimum of $8,000 and a maximum of over $160,000. By 2011, the average median income increased to $66 thousand (equivalent to the “orange category”) with a minimum of $13 thousand and a maximum of over $200,000.

Median Income in D.C., 1999/2011 (via ME!)

Median Income in D.C., 1999/2011 (via ME!)

For the most part, we would expect incomes to increase over time and typically they do, unless there’s a depression or recession (which there is, of course), but there’s inflation as well. While I don’t have the skills to convert the 1999 median income to 2011 inflation-adjusted dollars (which apparently are what the ACS numbers are), I can provide the compound annual growth rate (i.e. the average growth each year from 1999 to 2011) to give an idea of how quickly each median income grew. Keep in mind, this assumes a nice even growth rate between 1999 and 2011, since we all know a recession started about 2008 that the following numbers are this high are telling.

The left-side map below depicts the data holes (the white areas in the right-hand map). They are due to two reasons. First, some census tracts didn’t report any median income (in 1999, 2011, or both). Second, I attempted to account for census tract boundary changes between 1999 and 2010. This is illustrated by the interaction between red lines (2010 Census) and dark grey lines (2000 Census). In most cases, 2000 census tracts were split into two tracts for the 2010 census (population growth being the most logical reason). In two cases, 2000 census tracts were combined in the 2010 census (either population decline or combining two lower populated tracts into one larger tract). The right hand shows the explosive growth in median income for the District’s two primary areas of gentrification, mid-Northwest and Capitol Hill. In these areas, annual growth in median income averaged over 5.0% a year between 1999 and 2011. For comparative purposes, the District wide compound annual growth rate is 3.8% (the medium red category). While most of the strongest growth is occurring in these gentrifying neighborhoods, the other story are the stagnating neighborhoods (experiencing less than 2% growth if any) in the northern areas of Northwest, Northeast, and some areas in Southeast. However, it should be noted that there are also pockets of strong median income growth in Southeast as well. In addition, I’ve also reproduced the earlier map depicting the changes in D.C.’s “White alone” community, for comparative purposes.

CAGR of Median Income, 1999/2011 (via ME!)

CAGR of Median Income, 1999/2011 (via ME!)

Concentration of D.C.'s Whites, 2000/2010 (via ME!)

Concentration of D.C.’s White Community, 2000/2010 (via ME!)

In this post, we’ve taken an economic geographic (rather than cultural geographic) view of gentrification in the District of Columbia. Previously, we saw that the Mid-Northwest and Capitol Hill neighborhoods have become more populated by census respondents claiming to be “White alone”, while the District in general lost substantial numbers of “Black alone” claimants. With the data presented in this post, we have solid anecdotal evidence that strong median income growth correlates with increasing numbers of “White alone” residents in gentrifying neighborhoods. In addition, we’ve also seen that while median incomes in the District generally increased between 1999 and 2011 the increases, unsurprisingly, were not evenly distributed geographically. While the strongest growth rates are associated with the gentrifying neighborhoods, large areas of the northern quadrants experienced more stagnated median income growth, while some areas in the Southeast quadrant experienced negative growth over the period.

I believe a worthwhile follow-up post on this will focus on assessing whether or not there is an actual statistically significant correlation between race/ethnicity growth and median income growth. Stay tuned!

Visualizing “Gentrification”: Reversing White Flight in the District

Z Geography is mixing up the posting a bit, in addition to commenting on the news of the day from a geographic perspective, I need an outlet to chase after my geographic flights of fancy (Euclidean of course!). Hence my silence the past few days, which was spent data gathering and processing for today’s post which presents some analysis on the shifting demographic of the District of Columbia’s resident population.

Tobler’s (first and only) Law of Geography states that near things are more related to each other than they are to things farther away. A corollary, of course, is that near things are more often to become similar to each other, over time, then are things farther away – barring unique circumstances. In analyzing the shifting demographic balance between White and Black Americans in Washington, D.C. this corollary is particularly evident, as we shall see below.

But first! The data. All of data used in this post comes from two of the decennial (that’s every 10 years) censuses released by the U.S. Census Bureau, 2000 and 2010. You can “find” this data on the American FactFinder portion of the Census Bureau’s webspace. Ironic quotes because “finding” this data requires a good amount of time, particularly if you’ve never used it before. A note on the data, in this post I analyzed aggregate population head counts at the Census block group level, which is at larger scale (i.e. geographically “smaller”) than the census tract level, but smaller (i.e. geographically “larger”) than the block. Why not just go for the block? Partially because the Census Bureau is required, by law (in most cases), to protect individual’s information from public disclosure. So say you’re the only person living in a Census block, the results would not be reported to the general public because someone could look at the Census and see that you make $50,000 a year, have three children, rent your home, and are American-Indian (or Alaskan Native). While the risk of running into this problem in the city is small, I went with the Census block group. For this post, I analyzed the aggregate numbers of people claiming to be Non-Hispanic/Latino and, either “White alone” or “Black alone”. The Census has a two-tiered system (you may have noticed in some employment forms especially from the government) that distinguishes between Hispanics/Latino(a)s and Non-Hispanics/Latino(a)s. Choosing one of these and you can also choose whether you are “White only” or “Black only” or some other ethnicity/race. What this practically means is that the U.S. Census divides those who are “White Hispanics” from “Black Hispanics” and “White Non-Hispanics” (Caucasian Americans) and “Black Non-Hispanics” (African-Americans). Typically practice, though, is to report the numbers of “Black alone”, “White alone”, and Hispanic/Latino, with the understanding that the first two are from the non-Hispanic category. The “alone” population distinguishes this group from persons of two or more races.

The District of Columbia was home to 572,059 residents according to the 2000 Census. Of that population, 340,088 claimed to be “Black or African American alone” (Black), 59% of the total population, while 159,178 claimed to be “White alone” (White), 27% of the population. Like most cities, these population groups tended to concentrate in certain areas. In the District, one can perceive an East/West divide with the city’s Blacks concentrated in the eastern and southern quadrants (Northeast, Southeast) and Whites concentrated in the Northwest. D.C. is broken into four quadrants centered on the U.S. Capitol building. The President’s address is 1600 Pennsylvania Avenue NW, so next time you want to blame the President for domestic policy, don’t, blame Congress – they’re literally the center of the D.C. political universe (which would be another post in itself). By 2010, the population grew to 621,321 people an increase of almost 50,000 people in ten years (5,000 people a year on average). But what’s more interesting is for what group’s this population increase occurred. Among Whites, the population increased to 218,422, comprising 35% of the population. Notice that the increase in the White population group is actually larger than the increase in the city’s population as a whole (59,244 additional Whites on 49,262 additional D.C. residents). While the population claiming to be White increased substantially, the population claiming to be Black decreased. In 2010, 308,617 claimed to be Black, 49% of the total population, a decrease of 31,471 people. Where did this small town’s worth of people go? That will be the subject of a future post (I’m guessing the Maryland suburbs). While other population groups also experienced population increases, the real story here is the seeming reversal of the “white flight” phenomena during the 1960s and 1970s. Then, Whites “fled” urban areas for the relative safety of the suburbs. Now, it seems, Whites are returning (though it is hard to say based on this Census data from where they arriving, it may be another city) en masse.

The series of five images below highlight the geographic implications of this demographic shift, we can see the expanding (geographically) population of Whites in the District of Columbia. As I noted above, this population isn’t distributing itself evenly across the urban landscape, it is clustering. And these clusters are on the periphery of existing block groups that have high concentrations of Whites. Each graphic compares results from the 2000 Census (left) and 2010 Census (right). In each, “White alone” is represented by green and “Black alone” is represented by purple. The first map depicts these two identities together in a dot density map with each dot representing about 100 people. We can clearly distinguish the East/West divide with concentrations of green in the west and purple in the south and east. However, a strong exception is the area around Capitol Hill, in the 2000 map it is a cluster of 7 or 8 block groups west of an unpopulated area spanning west-east. This area, incidentally is the National Mall and encompasses the Capitol (the circular area in the far-eastern end). The increase in the “White alone” population is particularly noticeable between the 2000 and 2010 censuses in the Capitol Hill neighborhood. If you look closely you can also perceive the “creep” of the “White alone” group in the western fringes of the northwest area between 2000 and 2010. This is the advancing “gentrification” line that is steadily marching east across the city.

Dot Density of D.C.'s White and Black Communities, 2000/2010 (via ME!)

Dot Density of D.C.’s White and Black Communities, 2000/2010 (via ME!)

The next two maps are only of the changes in the “White alone” population and more clearly shows the geographic spread of that population group, notice the increasing footprint in the Capitol Hill neighborhood and the westerly advance in mid and upper (northern) Northwest. A good example neighborhood of this area is Columbia Heights. When I first moved to the District back in 2001 (just after the Census), I was told to be careful in Columbia Heights at night (one park in particular) and I almost never went. I’m happy to report that I still don’t really go, but when I went recently (2010 or so) the neighborhood had completely changed. There’s now a Target and a Best Buy and rent is topping $2,000 per month for a 1-bedroom. Of course, there’s still a number of Blacks still living in the neighborhood but there has been a noticeable increase in Whites. The first set of maps uses proportional symbols to illustrate the growing population in absolute numbers. The second set, a standard choropleth of graduated colors, compares the ratio of those claiming to be “White alone” in a block group with the total population of that block group.

Concentration of D.C.'s Whites, 2000/2010 (via ME!)

Concentration of D.C.’s White Community, 2000/2010 (via ME!)

Proportion of D.C.'s White Population, 2000/2010 (via ME!)

Proportion of D.C.’s White Population, 2000/2010 (via ME!)

The final two maps below depict the corresponding contraction of the “Black alone” population in the city. Again, the western fringes (in 2000) and Capitol Hill saw tremendous decline in population by 2010. Some block groups that, in 2000, over 76% of the residents claimed to be “Black alone”, decreased to between 51% and 75% “Black alone” a decade later. This is especially evident in the choropleth map of population ratios, note the northern and north-central areas of the city. The other interesting find is the decrease in the proportion of the “White alone” community in west-north-west. Without a corresponding increase evident in the “Black alone” community, it appears that other population groups (perhaps Hispanics or Asians) have grown substantially in this cluster.

Concentration of D.C.'s Black Community, 2000/2010 (via ME!)

Concentration of D.C.’s Black Community, 2000/2010 (via ME!)

Proportion of D.C.'s Black Population, 2000/2010 (via ME!)

Proportion of D.C.’s Black Population, 2000/2010 (via ME!)

I’ve purposefully tried not to make any qualitative judgments about the nature of this demographic shift, because its good or bad depending on your perspective. On the one hand “gentrification” is associated with beautifying communities, improved public schools, more eclectic businesses, general economic development and improved living standards, and reduced criminality. While I haven’t presented these datasets yet, most exist in the same place I gathered this data. But, more to the point, “gentrification” is also associated with pricing out the original inhabitants of communities. As land values increase, so do rents and taxes leaving the original residents, who could afford to live in these areas previously, are left in the lurch and forced to move.

In this post I’ve briefly touched on the thorny urban geography subject of gentrification. I’ve emphasized the race/ethnicity aspect of the phenomena by looking at the shifting boundary of the White and Black communities in the District of Columbia. We saw the shifting westward “gentrification line” and the geographic and demographic changes occurring in the Northwest quadrant and Capitol Hill. Of course, to present “gentrification” as simply a “White alone” activity is a poor characterization of a complex process. In a future post, I will explore the issue of gentrification within the Black community of D.C. itself.

White (British) Emigration from London

A BBC article from a month ago has been sitting on my tab for almost a month now, time to discuss! First, let me caveat by saying that the perspective I’m taking here, white British departing London is dictated by the article, not the data. Were I to do a more thorough look I would also like to know who is moving into these areas. So keep in mind this is only half the story. Overall I like this article and slick use of graphics, but I find its message a bit propagandistic – its not a story of “white flight”, “its a story of aspiration. A story of success.” Give me a break. I agree that we might not be dealing with “white flight” in the textbook definitional sense. According to, white flight (coined sometime between 1965 and 1970) referred to the movement of whites, especially middle-class whites, from neighborhoods undergoing racial integration. A more general definition is the movement of whites from areas where non-whites are settling. The second definition has the added quality that we can use it outside the context of a civil rights movement, since “racial integration” is obvious allusion to that. However, the article itself doesn’t really discuss the concurrent settlement of non-whites in those boroughs of London from where whites are departing so I think its difficult to call this not white flight by only providing half the data.

At any rate, there are additional points the article mentions that are worth repeating. First, and this isn’t really discussed much in the article, is the finding that while the British white population group is mostly declining in the “poorer” suburbs surrounding the central city, the proportion of whites is increasing in Inner London (coded blue in the map). This should sound familiar to my American readers, commonly called “gentrification”, the increase in urban whites has been noted since the late 1990s I would suppose. In Washington D.C., “immigrants” (who arrived in the city decades ago) are the ones moving to the suburbs as economics and family life permits. Taking their place in the city are richer, younger, white families. I would imagine that a similar process is ongoing in British cities. The article hints at this with a qualitative study of one surburban London neighborhood (Barking and Dagenham) noting that whites have left and replaced by black African migrants.

Another point worth highlighting is the depth of understanding the Barking and Dagenham borough. The article provides a political, economic, and social history of the neighborhood, which provides a sense of where and when it came from. This is a great example of the Kantian view of geography and history being intricately linked. Of course, the article does take the analytic leap with this history. It points out that the borough had a natural economic engine and lifeline, a Ford Motor Company plant, that subsequently closed. Now that whites are leaving and the main economic engine gone… well, what do you think will happen? More people will have commutes and for larger immigrant families with children finding care will become difficult among single families homes, though many may have relatives also living at home. There’s also the little variable of the poor economic environment, as a city well-integrated into the global economy (if not the most integrated city) its population is most susceptible during the global recession. Of course, not all population groups are created equal and the poorest classes in London are the ones most at-risk of un- or under-employment. Sadly, the article doesn’t take any of this into consideration.

The article gives the impression that these white Britishers moving away are primarily older families. Most bought their “council estates” (in the United Kingdom and Ireland this is a form of “public housing”) as they were entitled to do at 30% of its market value. They’re now flipping (to borrow the American term) these three-bedroom houses (selling at market value, presumably) and using the proceeds to buy much larger estates in the countryside. Again, what is left unclear is if these council estates are still, in fact, public housing and thus being sold to the new “owners” who pay 30% of the market rate while the government picks up the rest of the tab. Or if they are now private houses (the article suggests the latter) being sold to immigrant families. If its the latter that’s a pretty bad break for the immigrants who now enjoy the “privilege” of paying substantially larger mortgages without government assistance, and earning the ire of lower class British whites for “taking jobs” and living off “government money.”

A final point isn’t unique for the European context but interesting for the U.S. This white flight in the United Kingdom isn’t really a “white flight” its a British white flight, while the example used was a borough witnessing black African migration, a large part of the overall story are “other white” immigrants, mostly from Eastern Europe. In a sense this is a repeat of the early/mid-1800s Irish and German migration to the United States. While these immigrants were “white”, and even acknowledged as such, they weren’t exactly welcomed. The picture below, a great candidate for a geopolitical cartoon post (but alas!), drives this point home perfectly for the American context. It compares the experiences of blacks enslaved in the American South (labels above the scale) with Irish immigrants in the American North. Of course, I can’t read the caption on the bottom so I don’t know what the cartoon is really saying is equivalent. Irish immigrants could own property, were free to move about, seek different employment, and earn wages. They even fought in the Civil War. American Blacks, not so much.

Social equivalence? American Blacks and Irish immigrants, ca. 1850s (via

So I wonder about this BBC article, the message is too rosy for reality. Sometimes its worth remembering what’s not said in an article and disregarding what’s said. Yes, this is partially a story of economic advancement and the attainment of the Anglo-American dream, but the article did not convince me that wasn’t also white British flight.