Reason #763 why Houston is prosperous.

Matty G has a short post up on the economy of Breezy Point, Queens and my first reaction is “right, this is reason #763 why Houston is so prosperous.”

Mostly it has to do with annexation. At one extreme end you have a city like Philadelphia. Philly isn’t all that bad of a place, but when you look at the massive growth of NYC and DC, you have to consider the city’s development trajectory to be a failure. Philly lost population during the nineties and was flat during the aughts, and the city largely coasts on the infrastructure of previous generations. Roadway expansion (e.g. double-decking the Schuylkill) and transit expansion (e.g. Roosevelt Subway, Swampoodle Connection) have both gone nowhere. Taxes are high, services are low, and what little growth has occurred mostly takes the form of cancerous exurban development which has consumed productive farmland without much housing to show in return. Detroit follows the same pattern.

But if you look at the city boundaries this all makes sense. The place is hemmed in on all sides by small boroughs and townships. In some directions you can go from Center City to out-of-the-city in less than four miles.

The fun continues outside the city boundaries. There are no big suburbs outside Philly; instead, counties are a bouillabaisse of boroughs and townships of a couple thousand acres each. This is about the size of your basic Sunbelt master-planned community, so it isn’t particularly surprising that the local governments function more like homeowners associations. Exclusionary zoning is the norm, and most residential subdivisions are required to include large swaths of “open space” which is mostly about maintaining the visual deceit that you live in the “country” and not a suburb.

Meanwhile, in the middle of the annexation distribution you have a city like Dallas. Dallas goes out about eight to ten miles, plus Far North D, which is like an extended middle finger of garden apartments sticking into the adjacent cities. An east-west trip across Dallas is about 20 miles. But while Dallas is in the middle of the “city annexation” distribution, DFW as a whole is hardly in the 50th percentile in terms of prosperity and quality of life.

There is a reason for this: Where Dallas ends, the mega suburbs begin. Arlington and Plano are respectable cities in their own right, holding about 360,000 and 270,000 people, respectively. Carrollton, Frisco, Irving, and Grand Prairie each clock in well above 100,000. Large suburbs are not always favorable to “urban” things like mid-rise and high-rise structures, or LRT and other rail transit. Frisco in particular opted out of the DART taxing area, using sales tax money to subsidize commercial development instead. But large suburban jurisdictions generally tend to have coherent transportation planning, well-developed park systems, and a variety of housing types, including multifamily and small-lot single-family.

Big suburbs also make more effective use of land. The easiest way to measure this is population density. Bellevue, Washington clocks in at 4000 people per square mile. Plano “sprawls” at 3800, Arlington at 3900. Irvine, California holds 3200 a mile, and has what is perhaps the most coherent bike network in SoCal, combining near-100% arterial bike lanes with continuous off-street paths.

Meanwhile, back outside Philly, Exton – a major Amtrak and SEPTA stop – contains fewer than 1400 people per square mile, while neighboring East Whiteland township contains fewer than 1000. Density scarcely improves as you get closer in. The major corporate centers and edge cities of King of Prussia and Plymouth Meeting straddle multiple townships which each contain fewer than 2000 people per square mile. By contrast, even Frisco – which is exploding in population and has annexed a lot of vacant land in anticipation of future development – is already at 1900 people per square mile.

Which brings us to Houston, the opposite extreme. Houston’s expansion knows no geographic or political boundaries. When other cities incorporate, it just goes around. Suburbs like West U, Bellaire, and the Villages become enclaves. Strategic annexations of roadways and other tracts of land extend Houston’s reach even further. A cross-section of strategic Houston annexations from Prairie View to Lynchburg measures 65 miles across. A trip from Willowbrook to El Dorado is forty.

When your city is this large, it leads to some interesting paradoxes. Houston’s nominal population density of 3600 people per sq. mile is surpassed by many of the enclaves. West U, in particular, is north of 7000. But Houston is scattered with pockets of density above 10,000/sq. mi, and portions of Gulfton and Gulfgate check in north of 20k. Houston’s 3600 is also almost exactly 30% higher than Phoenix, which pursues the exact same transportation and annexation policies but with Euclidean zoning.

Really rough back-of-the-metaphorical-envelope calculations tell me that if you drew a line at Beltway 8, you’d come out with a population density in the 6000s. (Note: see followup post.) Which is really incredible. That’s higher than Portland or San Jose, and almost to Minneapolis and Seattle, both of which have more constrained geographic boundaries. And this density is achieved in a relatively young, Sunbelt city, that grew up almost entirely around cars.

Not bad.

The reader may note that the title of this blog post said “prosperous,” while an earlier paragraph on Dallas mentioned “quality of life.” This because the two are indirectly related. Prosperity is inextricably linked to population; you need people to have an economy, you need people with skills to have clusters, the more people you have the more skills there’ll be and the more clusters you’ll get. Likewise, life is typically more enjoyable if there’s more stuff. What the stuff is doesn’t matter – it might be restaurants, museums, churches, or death metal. People do stuff and the more people you put in reach of yourself the more stuff there is and the more likely it is you’ll find stuff you like.

Now there’s two ways to gain access to people and stuff. The first is you can put people closer together. The second is you can build faster transportation so it’s easier to get to them. The Chicago “L” is a slow loris, but at 11000 people per square mile citywide, it’ll still take you to a lot of places. The intersection of North and Halsted affords a view of a set of three successive 10mph corners on the Ravenswood and Evanston Ls. But as you can see there’s also a whole of stuff built there. On the other side, Phoenix is not particularly dense, but you can always hop on this thing and go wherever. Of course the best option is to combine density with high-speed transportation infrastructure which gets you the spur through Midtown or perhaps Roppongi.

Both of these things, land use and transportation, are more easily accomplished in a larger governmental jurisdiction. The primary opposition to land use is NIMBY – “I don’t want to look up at the Ashby high-rise while I’m mowing my lawn.” The primary opposition to transportation infrastructure is, again, NIMBY – “I don’t want to live next to this freeway.” The latter is somewhat more understandable, since freeways generate noise and pollution externalities that residential towers don’t. But in both cases you’re pitting NIMBY concerns against regional concerns. The larger a city you have, the more diluted the NIMBY voices are within the overall governmental framework.

About the only way to screw this up is to devolve decision-making authority to sub-units. DC and NYC do this with zoning decisions, which is sort of exactly why DC and NYC have ludicrous zoning policy. By contrast, Houston’s super neighborhoods are strictly advisory bodies. If you had any doubts you can check out the official site which uses the words “stakeholder,” “plan,” and “priority” twice each in the span of three paragraphs. Anytime you see those words, you know that no actual, real decisions are being made. And that’s the right way to do it.

13 thoughts on “Reason #763 why Houston is prosperous.”

  1. The larger a city you have, the more diluted the NIMBY voices are within the overall governmental framework.–this is not necessarily true, as some voices are more powerful–louder–than others. Even in Houston.

  2. This is necessarily true.

    Those more powerful/louder voices would be even more more powerful in a smaller city.

  3. Great vision by the annexation founder Oscar Holcombe. We are very “free market” type people and if someone wants governmental controls over what you can and cant do with your real estate, move to Dallas or Austin

  4. Boston is even worse. The city is geographically small, and fractured into tiny neighborhoods that make it difficult to do anything. Leave the city and you’re into small municipal fiefdoms. One of the nice things about LA (both the city and the county) is that they are large enough, in terms of geography and population, to actually be able to move on their own and make a difference.

  5. One of the differences between Phoenix and Houston is that there’s North Gateway Village and Desert View Village, both are which vastly underdeveloped and mostly desert. By contrast, Houston has the Addicks Reservoir and the George Bush Park, which are two large, undeveloped areas in town but much smaller than the deserts in Phoenix’s city limits; therefore, the reason why Houston is denser is because Phoenix has more undeveloped desert–not zoning. Take a look at Google Maps with city boundaries and you’ll see what I mean.

  6. Surprised why you didn’t cover Detroit. The Detroit population has been imploding for years, but it does have a few unique features: for an Eastern city, the city limits are huge–Manhattan, San Francisco, and Boston could all fit rather comfortably within it, and it still has a density of about 6500 per square mile, down from 13,000 in the golden years.

  7. Not sure how you’re estimating your figure of 6000 people/mi^2 inside the Beltway (or maybe I’m misunderstanding you), even if it is a rough calculation.

    According to the 2010 Census, there are 469,051 people living inside the Loop, and 1,597,359 between the Loop and the Beltway, for a total of 2,066,410 people living inside the Beltway. Roughly, the area inside the Beltway is 535 square miles, giving an population density of 3,862/mi^2.

    This is less than the 4,288 for the City of Portland and far less than 5400 for San Jose. For the ” Inner-Belt” to be as dense as San Jose is, it would need an extra 825,000 people living there. This undermines the assertion that the Inner-Belt is particularly dense.

  8. I think your numbers are probably pretty close to reality. When I wrote “inside Beltway 8” I was intuitively thinking “developed area inside Beltway 8,” which would be 610 plus a broad irregular arc that includes South Park, Sharpstown, Long Point, Acres Homes, and Kashmere.

    What I really would like to do is use the 2010 Census Data to determine a “maximum geographical extent” of a number of sub-Houstons of varying densities. But none of the census data that I’ve been able to download has the area of the tracts or blocks, so I can’t compute density with any kind of accuracy.

  9. Your annexation thesis was advanced by urban scholar David Rusk back in the 90’s in at least two books–Cities Without Suburbs and Baltimore Unbound. It makes sense so long as you assume the city center is poor and the suburbs are prosperous. That’s been the American pattern for decades, but it’s not the pattern in the rest of the world. And there’s fairly strong evidence that at least in some cities–Boston, Washington, San Francisco–the center is becoming pretty prosperous. If that’s the case then a city doesn’t need to massively annex to be prosperous on a per capita basis.

    For your example of Philadelphia (and similarly situated cities), however, there aren’t enough “gentry” in the city center to balance out all the poor people in the second ring. Then the suburbs around it really do act like a choke collar–like Chicago’s name for the suburbs, “collar counties”. Combat over resources occurs in regional agencies, among other places.

    It does seem that cities that are large in absolute population can have more activist governments. New York does a lot of things that other cities don’t, caused by more than just the fact that 5 counties are contained within it. Los Angeles isn’t quite on New York’s level but still has a lot of capacity. I am a supporter of activist cities–depending on what they’re doing–but of course not everybody is.

  10. I don’t think “prosperity on a per capita basis” means anything. Warren Buffet and Jeff Bezos and Tim Cook could all move onto the same farm and incorporate, but that wouldn’t be an actual city, it’d just be a farm with a bunch of rich guys on it.

    In the case of SF there are a lot of very high-income jobs there which means there are a lot of very-high-income people, but a substantial chunk of that money is just poured right back into inflated housing costs which just accrues as rents to incumbent landowners. If you kept SF’s high-tech cluster but reduced the cost of living to Houston levels you’d have a lot more cash floating around which would then be reinvested in all sorts of things, new business ventures, new developments, what have you. The place would explode.

Comments are closed.