If we’re in an urban renaissance, why are cities still losing population?

Despite the general feeling among urbanists that the city is making a comeback after half a century of neglect, I still read from a lot of suburbanists (a catch-all term I’m using to describe Joel Kotkin, Wendell Cox [see comments], etc.)—and even the mainstream media—that cities are still losing population. I don’t have a lot of patience for statistics, so it all becomes a sort of he said/she said argument to me, but here’re a few opinions from the pro-“cities are becoming more desirable” side.

First, here’s Ryan Avent, who argues that looking at population stats is misleading without taking into account prices:

But of course, population growth is an unreliable indicator of demand, because of the all important supply side of the market. Imagine two areas: Gotham and Pleasantville. Say the demand to live in Pleasantville increases a little while the demand to live in Gotham soars. And say that due to differences in land use restrictions, housing supply responds dramatically in Pleasantville and very little in Gotham. Then what we’ll observe in Pleasantville is a rapid increase in population and slower growth in prices, and what we’ll observe in Gotham is rapid growth in prices and slower growth in population. And this is exactly what we have observed in the real world. Suburbs have seen massive housing growth and rapid population growth, but prices in central cities have soared, even in many places where population numbers are level or falling. If no one wanted to live in central cities, prices for homes there would not rise. And indeed, several decades ago, prices for homes in big central cities were dropping. But that trend has clearly reversed. You can’t draw conclusions about demand shifts from population numbers alone. This is a very simple point, and yet its repeatedly ignored.

And here’s Michael Lewyn (with a new blog! …which unfortunately has no RSS feed. Can anyone get in touch with him and ask him to rectify that?), arguing that we need to look at where in these declining cities the growth is happening, using St. Louis as an example:

Big winners: the most urban parts of St. Louis.  St. Louis’s once-desolate downtown gained population in huge numbers, and near-downtown areas like Soulard, Lafayette Square and the Central West End gained population. The losers?  The bad neighborhoods on the North Side and the more pseudo-suburban areas in the South Side. So what does this tell me?  That even in St. Louis, a city that has lost population faster than Detroit over the past few decades, the MOST urban areas are gaining. Even in St. Louis, people want to live in the city in greater numbers each year, as long as the environment is fairly city-like (i.e. downtown or close to it). The only thing that hasn’t happened is that gentrification hasn’t trickled down to the less urban neighborhoods further out.

He also makes the same point about Chicago.

And finally, Tim Evans on a blog about smart growth in New Jersey, who points out that Kotkin’s definition of a “city” leaves a lot to be desired:

Only five New Jersey municipalities make the list [of New Jersey cities], three of which (Edison, Union and Wayne) are more like the kind of places most people, probably including Kotkin himself, have in mind when they think of the suburbs.  (Wayne, incidentally, is the hometown of — and inspiration for — the unabashedly suburban rock band Fountains of Wayne, who frequently sing about life in the North Jersey suburbs.)

More tellingly, consider some North Jersey cities that are NOT on the list:   Paterson (New Jersey’s third-largest city), Elizabeth (fourth-largest), Clifton, Passaic, Orange, East Orange, Bayonne, Perth Amboy, Plainfield, Hackensack, Linden and Rahway.  Not to mention Union City, West New York and Hoboken, three of the most densely populated municipalities in the entire United States (all three exceed 30,000 people per square mile).  Not to mention a whole host of smaller but still “urban” places such as Asbury Park, Keansburg, Neptune, Long Branch, Garfield…

Oh, and Jersey City, the 2nd-largest city in the state.

All of these places fail to meet one or the other of the four criteria for qualifying as a “principal city.”  In other words, they are all “suburbs” by Kotkin’s definition.  That’s right, Jersey City is a suburb.  In fact, so are 23 of the 30 “distressed urban communities” identified by the Housing and Community Development Network in its Cities in Transition report of 2006.

So, there you have it. Seems pretty damning to me, but if Joel Kotkin or any suburbanists are out there reading and would like to respond to their critics, I’d be more than happy to post their response.

The only thing that I’d say in response is that (sub)urban trends seem to take a very long time to play out. From what I understand, cities were still growing in the ’40s and ’50s (and maybe even early ’60s?), despite the fact that many of the pro-suburban policies (height restrictions, banning of new elevated rail projects, zoning, parking restrictions) started going into effect soon after (if not before) the turn of the century. Given that many cities haven’t even gotten around to parking reform, I’d wait a few decades before declaring the urban renaissance a sham.

  • http://myurbangeneration.wordpress.com/ Eliza

    forwarded to Lewyn

  • http://myurbangeneration.wordpress.com/ Eliza

    Cox didn’t actually say (most) cities are losing population. He just said they were gaining it slower than the suburbs. Which I think even most urbanists wouldn’t contest. Last I heard Detroit was happy that it’s bleeding has slowed even if it hasn’t quite reversed. Headline: Even Cox admits (most) cities are growing! Yay!

  • http://homepage.mac.com/jjakucyk/ Jeffrey Jakucyk

    Ryan Avent’s points are spot on. It’s no secret that housing tends to be more expensive in the city, but few people stop to ask why. There’s some intrinsic factors, like that the infrastructure is no longer brand new so maintenance can’t be deferred anymore, landscaping is more mature which increases value, city residents are taxed for services that are used by suburban residents who don’t pay for them, such as main libraries, wider surface streets to handle more traffic, etc.

    The main factor though is that there’s demand for good urban living and not enough supply. Note that I said *good* urban living. Pseudo-suburban areas, as Michael Lewyn mentions, are in a lot of trouble. Even many inner-city highly urban areas can have issues that aren’t intrinsic to their built form. Crime and bad schools are all it takes to send an otherwise great neighborhood down the toilet. If those issues could be addressed easily, we’d see a massive urban renaissance because it would solve much of the supply problem. With demand and supply more even, it would then let neighborhoods turn around gradually, without cataclysmic gentrification.

  • http://lewyn.tripod.com/blog Michael Lewyn

    Thanks for the plug for the (actually not new, just revived) blog. Am currently too technically retarded to create RSS feed, though am amenable to advice.

  • awp

    Avent’s right, Demand and Supply are closely related to both quantity and PRICE. It’s like saying Fords are better cars than Roll Royces just because Ford sells so many more.

    Also, already dense areas will not grow in either percentage or aggregate terms as fast as less dense areas, due to legacy/capital costs among other factors, even in Houston where you are actually allowed to bulldoze old residences and put up new ones.

    Also, definitions matter.
    The main fact in this whole debate is that Metropolitan Areas/urban areas are growing faster than Rural areas. People are moving to “Cities” more properly defined (generally urban area as defined by the census or metropolitan area as defined by the office of management and budget). I stop paying attention to the people arguing that suburban areas are growing faster urban areas (even though I think it is true) when I realize that, most of the time, they are calling anything other than the central city within an Metropolitan area a suburb. To my knowledge The City of Houston is the closest we get to a central city that actually encompasses a significant part of the “urbanized area” within its urban area.

  • http://rationalitate.blogspot.com Rationalitate

    It looks like the problem might just be that it’s hosted on Tripod—I’m not sure if they offer the option to create a feed. I know it would be a pain, but you might wanna consider moving to a more modern host (Blogspot, WordPress, etc.). It will probably be difficult (if not impossible) to move all your old posts, since Tripod is such an old hosting service, but even if you have to start over, it’s definitely worth it—without an RSS feed, you have to rely on people remembering to check the site constantly, and it ends up being more of a website than a blog.

  • Emily Washington

    Ed Glaeser gives this topic great treatment in his new book. He explains that in areas with the most stringent height restrictions, populations are likely to shrink as larger, poorer families move out and wealthier, smaller families move in. He offers the example of historic areas in Manhattan where density essentially cannot increase and where residents are 74% wealthier than other Manhattan residents (p. 150).

  • Alon Levy

    Right… but he makes a blanket claim, even for areas where the central city outgrows the suburbs, like New York. (Or maybe I’m confusing what he said under his own name and what Mixner said in comments. I’m not sure right now.)

  • http://twitter.com/EatUrLawn Gregory Wade

    Uh…Landscaping is Commercial.

  • http://twitter.com/EatUrLawn Gregory Wade

    Frankly, I don’t see this as a matter of he said/she said, but a matter of demographics. Demographers expect Developed Countries to generally experience declines in Urban populations and increases in Rural populations; a reversal of 20th century trends. On the other hand, Developing countries are expected to experience the reverse. The suburbs are merely an Urban appendage.

  • http://homepage.mac.com/jjakucyk/ Jeffrey Jakucyk

    What do you mean by that? My only point was that in established neighborhoods the trees and other landscaping is fully grown. That makes the area more valuable than in new greenfield developments which usually don’t have much appreciable landscaping beyond grass and a few spindly saplings.

  • awp

    Really? Do you have a link to where a respectable demographer has claimed that people are going to move away from Metropolitan/urban areas to rural areas? Cause if you do and he/she is right, he has the best crystal ball ever, because that is not the way population is trending.

  • http://lupussolusluna.blogspot.com/ LoboSolo

    Not so fast there urban cowboy. Parts of the urban areas are growing and gaining value while other parts are in decay. It’s often trendy (and cheaper) for some to renovate old houses in some neighborhoods than to buy a new house in the ‘burbs.

    And the demographics show that most of the urban growth comes from immigrants while more Americans tend to move to the suburbs.

    For those of us who do enjoy urban living, it is often difficult to convince our co-workers that we aren’t afraid for our lives by living in the “crime-ridden” city. I like have a small house with a small yard in the city … which ironically … 80 years ago was itself a suburb rather than a large house with a large yard that I must maintain.

    To each his own …

  • Anonymous

    Its occurred to me that there is a serious shortage of urban options in the US. Thats why prices in the few cities that remain is so high. Years ago there were many more cities to live in, but now they are mostly either decayed or destroyed and replaced with suburbia or godawful projects. Smaller cities in the US mostly lack the amenities they had 60 years ago, which now can only be found in the bigger, and vastly more expensive cities.

    My highest hope is that the “urban renaissance” can turn around and re-urbanize some of these smaller cities, so cities like New York and San Francisco can have their prices come down a bit.

  • Jim654

    The new census data shows that suburbs continue to grow much faster than cities. And yet we’re supposed to believe that there has been, in Ryan Avent’s words, a “sharp reversal” of the longstanding trend in favor of suburbs. Sorry, it’s ludicrous.

    As for prices, housing prices tend to be higher in dense urban areas not because of higher demand for such housing but because dense housing costs more to supply. Land is more expensive because more people are competing for each square foot and because, in mixed-use development, homebuyers are competing for land with commercial enterprises like stores and offices that can afford to pay more. Construction costs are higher because it is more expensive to build vertically than horizontally. And housing prices have increased in some dense urban areas not because of a shift in demand but because of gentrification and population growth through immigration and natural increase.

    There is clearly a market for dense, urban housing. But it is a small, niche market. As a share of the total housing market it continues to shrink. One wonders just how much more overwhelmingly dominant the suburbs will have to become before “urbanists” let go of their fantasies and acknowledge this reality.

  • http://alexblock.net Alex B.

    Anyone (especially Kotkin) attempting to answer this question without first being forthright about their definition of what is a ‘city’ and what is a ‘suburb’ shouldn’t be taken too seriously.

    It’s a problem of definition. This isn’t a new issue, either. There are plenty of suburban areas within central city jurisdictions. There are plenty of urban suburb-level jurisdictions as well. Municipal boundaries are not particularly useful from an analytic standpoint.

  • http://www.UrbanCincy.com/ Randy A. Simes

    Population growth is a bad benchmark for older cities for another major reason – household sizes have decreased relative to the amount of square footage demanded. What this means is that a densely built section of Chicago, Philadelphia, Cincinnati, New York, Boston, Cleveland or Pittsburgh could boast 100% occupancy, but still experience population decline. Those densely built row houses may now only house 5 people when at one time they housed 20. In both scenarios the structure was 100% occupied, but over time cities have become less densely populated.

    You speak to this with the land value comparison. Since there is increasingly higher occupancy rates, land values are increasing even while populations are lower when compared to past years. It’s a pretty simple lesson, but one where people must examine the history’s of our cities and understand how that history differs to what is happening today. It appears as though Kotkin cares to ignore such details.

  • http://rationalitate.blogspot.com Rationalitate

    So wait, you say it’s “not because of higher demand,” but rather because “more people are competing for each square foot”? How are those not the same thing?

  • awp

    Austin Contrarian has a post on this phenomenon in regards to central Austin.

    http://www.austincontrarian.com/austincontrarian/2011/03/78704-and-the-2010-census.html

    He seems to have a series going analyzing the relative changes of census tract populations across Austin

  • Anonymous

    “infrastructure is no longer brand new so maintenance can’t be deferred anymore”: seems to me this would lower prices rather than raise them (would you pay more for a house with a $5K annual maintenance bill, or a similar one with a $500 annual maintenance bill?)

  • Anonymous

    “infrastructure is no longer brand new so maintenance can’t be deferred anymore”: seems to me this would lower prices rather than raise them (would you pay more for a house with a $5K annual maintenance bill, or a similar one with a $500 annual maintenance bill?)

  • http://homepage.mac.com/jjakucyk/ Jeffrey Jakucyk

    It keeps prices lower in the new suburbs because they don’t have to do any maintenance on their infrastructure since it’s all brand new. In cities, the infrastructure is older and is in various stages of decay, repair, and replacement. So while the cost of the new roads, sewer, and other utilities are generally built in to the sale price of houses in new subdivisions. However, their tax bill doesn’t reflect the maintenance costs that will be required in 20-30 years and beyond. They ARE reflected in the tax bills in cities and many inner-ring suburbs. The same is true for the externalized costs of new suburban development, such as widening roads outside the subdivision, increasing capacity for water and sewer treatment plants, more schools, libraries, expanded police and fire protection, etc. There’s a lag in implementing these services that people usually demand, so taxes have to go up over time to pay for them. This happened ages ago in already developed areas though. Aside from all that, there is of course the fact that land is simply more valuable the closer to the center of the city you get. If zoning precludes increasing density, which distributes that same cost among more people, then of course that single-family house in the city is going to cost more than the same house in the suburbs, just because it’s more convenient to jobs, shopping, and attractions.

  • Anonymous

    I agree with what you say, but that only explains why taxes would be higher in the city, not housing prices. In fact, high taxes should lead to lower prices (comparing City A with $5K a year in taxes and City B with $2K a year, all else being equal, a person would be willing to pay up to $3K more to live in City B).

    If you do a total-cost-of-living type calculation, then it makes sense to looks at taxes and such, but most people don’t– they just look at “a house in this neighborhood is $800K and a house in this neighborhood is $500K”.

  • http://homepage.mac.com/jjakucyk/ Jeffrey Jakucyk

    Well yes, but as I also said, as have others, the higher prices in the city are a reflection of higher demand and not enough supply. There’s other things that are more difficult to quantify too. If you took two otherwise identical neighborhoods, and one has a mature tree canopy and the other doesn’t, the one with the trees is going to command a higher price. Cities and established suburbs generally have this advantage over brand new suburbs, simply due to age. Also, buildings in pre-WWII areas tend to be much more substantially built than they are today. You can find great architecture and nice built-ins, significant woodwork, etc., in older buildings. While it certainly can be built today, it’s generally not being done even in pretty high-end markets. So those who really value good design, quality, uniqueness, aren’t interested in suburban housing subdivisions. It’s another case of supply and demand, and the supply is mostly in the city. There’s also things like walkability, proximity to nice shops, established parks, recreation and cultural facilities, etc., that tend to be missing in suburban areas and thus suppress desirability and values somewhat.

  • Anonymous

    Most housing in core cities is quite old, so the cost of building vertically is paid off. The cost of building vertically is also far lower than the present cost differential between cities and suburbs. Third, most cities are not Manhattan and the Vancouver core. Even within cities where high rises are popular like these, they are a minority of all housing.

  • Jeff in LA

    I worked for the US Census Bureau in Central Los Angeles last year. Census Bureau management hired about 70% of authorized staffing, failed to distribute census materials in a timely manner or in accordance with published manuals, failed to hire sufficient numbers of foreign-language speakers sufficient to conduct canvassing effectively, and yet demanded that Census results be generated just as quickly in poor, high-density neighborhoods with large numbers of non-English speakers as in affluent suburban neighborhoods. And we wonder why Census numbers show relatively small amounts of growth in cities?

  • http://urbanplacesandspaces.blogspot.com Rlaymandc

    First, I think you’re right that it takes a long time to play out. Second, the reality is that the “growth” is uneven. Strong population inflow at the core of a center city, and continued leakage in other parts of the city, except if there is a strong program of intensification at transit centers. This needs to be addressed big time. Third, most center city real estate markets are weak even so, so what’s happening in Washington, DC is an outlier. Fourth, the population increase is a function of attracting younger households (although the comment by Randy Simes about household sizes is important) and there is a big problem with retention as household type changes–from single to married, from married no children to married with children–and amenities demands and other requirements (e.g., quality local public schools) can’t necessarily be met by urban locations. Fifth, of course the continued problem of leakage of employment from most center cities provides a strong push for suburban residential location because of work proximity commuting issues.

  • http://urbanplacesandspaces.blogspot.com Rlaymandc

    First, I think you’re right that it takes a long time to play out. Second, the reality is that the “growth” is uneven. Strong population inflow at the core of a center city, and continued leakage in other parts of the city, except if there is a strong program of intensification at transit centers. This needs to be addressed big time. Third, most center city real estate markets are weak even so, so what’s happening in Washington, DC is an outlier. Fourth, the population increase is a function of attracting younger households (although the comment by Randy Simes about household sizes is important) and there is a big problem with retention as household type changes–from single to married, from married no children to married with children–and amenities demands and other requirements (e.g., quality local public schools) can’t necessarily be met by urban locations. Fifth, of course the continued problem of leakage of employment from most center cities provides a strong push for suburban residential location because of work proximity commuting issues.

  • http://urbanplacesandspaces.blogspot.com Rlaymandc

    First, I think you’re right that it takes a long time to play out. Second, the reality is that the “growth” is uneven. Strong population inflow at the core of a center city, and continued leakage in other parts of the city, except if there is a strong program of intensification at transit centers. This needs to be addressed big time. Third, most center city real estate markets are weak even so, so what’s happening in Washington, DC is an outlier. Fourth, the population increase is a function of attracting younger households (although the comment by Randy Simes about household sizes is important) and there is a big problem with retention as household type changes–from single to married, from married no children to married with children–and amenities demands and other requirements (e.g., quality local public schools) can’t necessarily be met by urban locations. Fifth, of course the continued problem of leakage of employment from most center cities provides a strong push for suburban residential location because of work proximity commuting issues.

  • Pingback: posez une question mail