Monday, October 3, 2011

Note on Gentrification and Preservation

There have been several great pieces on the dynamics of gentrification over the past several days. Market Urbanism's Stephen Smith, who is now writing at Forbes, observes how permitting more development in desirable areas will help "soften the blow of gentrification." Alon Levy theorizes about the causes of central city revitalization. Emily Washington, in response, considers the time preferences of the upper and middle-classes. At the City Paper, Lydia DePillis discusses the relationship between historic preservation and gentrification, among other things.

In DePillis' article, in particular, there is this observation linking preservation and neighborhood revitalization:

"The thing is, preserving distinguished architecture and well-constructed neighborhoods can be one of the best ways for real estate to hold its value. Places that look like they were once prosperous send a message that they could be again."
However, one can argue that it is well-constructed buildings and distinguished architecture which attracts gentrifiers long before historic districts are contemplated.  The following chart was put together some years ago for a project looking at four southern cities (Nashville, Atlanta, Memphis and Savannah) in which all census tracts in 1940 were grouped into quartiles based on housing value in that year.  The tracts were then revisited in the 1970 and 2000 census years, keeping the groupings the same (the numbers show quartile housing value divided by the median housing value for the city in a given census year):

The pattern over time distinctly shows the operation of residential "filtering" from 1940 to 1970, as homes in wealthier neighborhoods came to be occuped by those from poorer neighborhoods, and an almost mirror-image reversal of this trend from 1970 to 2000 (intervening census years show that the turning point is in fact around 1970).  The middle class was simply returning to the same homes that their parents' generation had abandoned some twenty to thirty years before.  In 1970, moreover, one could purchase a home built by, say, an upper middle-class family of the 1920s, for what was essentially a 50 percent discount a house that went for four times the price of those in the poorest quartile in 1940 was, in 1970, selling for only twice that, even though these were in most cases the very same houses. 

This is different result than some other studies (see e.g. Old Homes, Externalities, and Poor Neighborhoods: A Model of Urban Decline and Renewal) which find that neighborhoods should cycle over time as housing stock ages, deteriorates and ultimately becomes ripe for redevelopment.  Yet the poorest two quartiles have not cycled at all in 60 years, while the housing in the first quartile was not actually redeveloped, but only renovated by new buyers (that is, its age was an amenity, perhaps the primary amenity, rather than a liability).  The hidden factor is the quality of the housing stock, which overwhelms the effects of aging and deterioration.

This can help explain the perceived attitudes of preservationists toward neighborhood change that DePillis discusses in her article.  It also suggests an explanation for the suburbanization of those with low incomes: as the better urban housing is gentrified, the former occupants may have two choices: move back into impoverished central city neighborhoods or move out into the aging suburban housing vacated by the gentrifiers or those moving to newer suburban houses. If costs are roughly comparable, the decision is easy. 

This type of gentrification is different from the the arrival of residents back into a formely commercial or industrial urban core, where the pattern is much more analagous to greenfield development (as Alon mentions in his post).  This is occurring simultaneously with gentrification of existing neighborhoods in Nashville and other cities, but takes a very different form. 


  1. I am currently researching preservationism and architecture. I am wondering if there is any correlation between the rise of brutalist modern architecture and preservationism. So far my research is mostly conjecture and coincidence but I can't help but think of the backlash some bloggers had over preserving some 1950s strip malls in DC. See:

    I have to wonder if modern architecture is too divorced from the tastes of the general public, and that this is contributing to the lack of redevelopment.

  2. James, people like Andres Duany and James Howard Kunstler like to say that the rise of the preservation movement came about when the new became worse than the old. We've been demolishing some pretty awesome buildings throughout human history, but it was always ok because the replacement was better.

    Once we got to the 1960s or thereabouts, this was no longer the case. Brutalist architecture is one example, partly because it's so disconnected from the tastes of the general public, but also because so many buildings were razed for parking lots or just empty fields.

    Aside from that, the standard development pattern of wide roads, parking lots, and sprawl in general means that many people not only would rather see an old building remain, but they'd rather a fallow field remain that see it get developed. If we'd actually build good buildings and nice places that people can relate to and care about, then we wouldn't be in such a pickle.

  3. Jeffrey,

    I have to wonder about that, because while I am not a fan of a lot of modern architecture I find myself asking "did people like Art Deco when it was new? Or have we merely grown accustomed to it?" There were NIMBY groups 100+ years ago (they blocked the Loop portion of Chicago's L for many years) but I have the impression that they became more powerful and vociferous in the 70s.

  4. James,

    I don't know what changes in attitudes toward architecture, but in one respect there was a sea change: in the 1950s and 60s the government at all levels condemned entire neighborhoods to make room for freeways and urban renewal projects. The scale of this was larger than that of any previous scheme. Conversely, many of the neighborhoods affected were fairly organized and had a middle class that could resist. The residents of Greenwich Village of the 1950s were much richer and more organized than those of Five Points of the 1890s.


    To be honest, I wasn't thinking of Southern cities, so I didn't think of the housing stock angle. In New York, it's not as much of an issue, especially not in the northern sector, which I'm most familiar with. But, at least the older prefabs have been improved upon: in the era before HOAs, home ownership meant that people continuously improved upon Levitt's mass-produced designs, so that now you can't even tell you're in a Levittown.

    But you're right that suburbanization of poverty should include an element of housing quality, and in metro areas that have seen significant growth in the last 50 years (certainly not New York), it is partly about old vs. modernist housing.

  5. Alon: Even though I only looked at southern cities, I think the basic principle should be applicable to almost any American city. Gentrifying neighborhoods are usually those which were once well-to-do, partly (mostly?) because the housing there is of high quality. Even in New York, Greenwich Village gentrified while the Lower East Side lagged. And even in the city's direst years, Greenwich Village remained wealthier than the LES since the housing stock was more desirable there. It is unusual for neighborhoods to leapfrog one another in economic status, although the spread may contract or expand.

    In Nashville, the consistently poor census tracts are about the same distance from the CBD as the gentrifying ones, have the same access to transportation, the same quality roads and infrastructure, and even the same poor schools at the outset, yet they did not gentrify at all. Drive through them and it's obvious that the housing stock is of a far lower quality than in the gentrifying areas. There are ramshackle shotguns and very small, cheaply-built bungalows. Those who had any money in the 1950s and 60s used it to trade up to middle class-built homes in the now-gentrifying areas as their owners left for the auto suburbs, not to upgrade their existing homes. There is constant fear that gentrification will spread to these areas, but the odds of that seem low to me, since the prerequisites for "existing stock" gentrification (high quality houses at comparatively low prices) simply do not exist there.

  6. On the northern margins of the UES and UWS you have areas that used to be run-down but are now indistinguishable from the main neighborhood. That's why I brought up citations for people treating the East Side north of 86th as bad in the 1960s, whereas now the boundary is 96th. The change in Morningside Heights is even starker - it went from a neighborhood of flophouses to the richest neighborhood in Upper Manhattan, leapfrogging Washington Heights and Inwood. And so far Harlem has not gentrified too much, despite having the same housing as Morningside Heights and Yorkville.

    The issue here, I think, is that New York has so much high-quality prewar housing that the line separating good housing stock from prefab crap is too far out to act as a color line. Instead, the natural boundary comes from commute convenience, i.e. the boundary of the subway system. In the Bronx this corresponds to city limits fairly well, but in Queens it doesn't and it's likely that eventually there will be a stark line in Queens between middle-class or rich neighborhoods near the subway and (especially in the ill-favored southern half of the borough) poor neighborhoods beyond the subway or in the Rockaways.

  7. Alon -- while it's true you'll see some leapfrogging, overall I would expect that the geographic distribution of relative incomes in Manhattan broadly parallels what one would have seen in 1950 or 1900. The LES and Harlem continue to be the lowest-income areas in Manhattan, which has been true for many decades (whereas the UES has been a wealthy area since the late 1800s).

    Now given the housing stock in Harlem and the LES I think we could theorize that, in time, these areas will gentrify too (that appears to be well underway for the LES actually). A complicating factor as you mentioned is the presence of public housing, which was built in large quantity in both the LES and Harlem (but less so in Morningside Heights), its location reflecting the geography of urban poverty in the 1950s and 1960s. That more than the housing stock may be slowing the process down in those locations (note that 96th St. marks the southernmost extent of the extensive public housing built in East Harlem).

    As far as commutes go, it would be interesting to visualize the regional trends over time -- that sort of trend might already be visible over the 1990-2000-2010 census periods.

  8. This is kind of off-topic, but I think it's of interest to everyone who reads this blog. It's an article about Jane Jacobs and some of her critics.

  9. Toronto has had some interesting gentrification trends. There are a few neighbourhoods that were very poor when they were built, but still had their charm and the homes were mostly small or crowded rather than poorly built (though still more simplistic than wealthier homes). They remained poor for a while when suburbs were more popular and the city was less desirable, but now that that's changed, partly due to long commutes and traffic, they've become popular. And while a 1000 sqft house might seem small, it feels pretty roomy compared to all the 400-800 sqft condos being built.

    Cabbagetown is one of the best examples, it used to be a poor Irish neighbourhood about 100-150 years old but would now be considered solidly 2nd tier. There is also gentrification occuring in the form of reconstruction, as is occuring in many parts of East York were small bungalows are being replaced by 2 and 2.5 story homes.

  10. Thanks for your comment, Anonymous. Based on the google street view, Cabbagetown appears to have a much higher quality housing stock than Nashville's poorer neighborhoods, and is considerably higher density as well (actually, higher density than any large-scale residential area in Nashville). Few American cities are fortunate enough to have a neighborhood like that! An example of the type of housing I am referring to would be the kind shown here -- dilapidated, small, poor quality and fairly low density to boot:,+tn&hl=en&ll=36.147366,-86.765733&spn=0.000002,0.001287&hnear=Nashville,+Davidson,+Tennessee&t=h&z=20&vpsrc=6&layer=c&cbll=36.147282,-86.765678&panoid=Dx3bmoaytLUBMsKIYVHZYQ&cbp=12,17.53,,0,1.88

    Small square footage is not necessarily a handicap, as you mention, but I think you'll need reasonably high intrinsic architectual and structural quality to compensate for it.

  11. Do you think those homes in Nashville could be improved through renovations? If not, I guess at least there are many vacant lots to build on, and the homes are on relatively large lots, which gives flexibility for rebuilding.

    Anyways, I'll try to find some more modest looking homes in Toronto. Starting in Cabbagetown, if you look at a streetview of Alpha Avenue, you'll find some small rowhouses that look very charming, but don't have the fancy architecture of some of the other homes in the neighbourhood. Granted, they have more potential than those Nashville homes.

    If you look at 62 De Grassi Street, you'll find a pair of cottages that might be smaller than even the smallest of those Nashville homes, but they're well kept and look nice. If you look at 122 Robinson Street, you'll find very similar cottages, but they're more downbeat looking, but the cottages just across the street look nicer, and there are more further East on the same street.

    If you look at 791 Craven, you'll see a stretch of very small houses built in the early 20th century in an area dubbed Tiny Town. The homes are in decent shape though, and some have been recently renovated and there are even a few new homes. For example, that strip of dirt where google shows 791 Craven (it's actually 781) is now home to this little prefab:
    This neighbourhoods will likely always be one of the cheapest in the inner city, but it still has potential for yuppies/small households, especially considering that they're cheaper and bigger than many of the new condos being built in Toronto.

    If you go to 403 Norcliffe Boulevard, you'll find more small homes, although these are a bit newer, maybe even early post-war. The neighbourhood, Oakwood-Vaughan looks nicer than Tiny Town though, with colourful front yard gardens, porches, the homes are in good shape, there's a few new homes too. The neighbourhood used to have a bit of a bad rep, but it's getting quite popular. It's home to a lot of Italians, Portuguese, Filipinos and some Latin Americans and Jamaicans and most of those homes are worth $300-600k.
    It's also home to the tinest house in Toronto:

    It's true that the neighbourhood you linked in Nashville will probably not be the first to gentrify. However, I think the potential for it to improve is there if there is enough demand for living near downtown. Being used to Toronto, Nashville looks pretty empty, and I think if those empty spaces get filled up it will already look nicer. Add in some flowers and porches, and I think it could have a similar feel of Oakwood-Vaughan.


  12. This comment has been removed by the author.

  13. I've always had a pet theory about the linkages between gentrification and the filtering of housing but have so far been unable to see this trend reflected in the data. I was trying to compare metro level existing vs new housing prices. When gentrification is happening there should be a closing of the gap between existing and new housing prices, as older architecture is becoming more valuable in gentrifying neighborhoods, but this simply wasn't the case wherever I looked. It's interesting to see it come across when looking at quartiles!