Austin Contrarian

City dwellers

The Census Bureau has, at last, begun to release urbanized area data. As Nate Berg notes at the Atlantic Cities, the headline that 80.7% of Americans now live in "urban areas" is a bit misleading. Or at least confusing. "Urban areas" consist of "urbanized areas" and "urban clusters." An "urbanized area" is a contiguous area of at least 50,000 people meeting certain minimum density threshholds. An "urban cluster" is, basically, just a small town -- too small to qualify as an urbanized area, but too densely settled to label "rural." That 80.7% includes a lot of people who live in small towns, far from any place one would call "urban."

If we want to count city dwellers, we should stick to urbanized areas. "Urbanized area" is the best fit for our intuitive notion of a "city." Urbanized areas have at least 50,000 people, which is a reasonable threshhold for separating "city" from "town." The urbanized area boundaries are determined by complicated algorithms that measure density and distance from other dense clusters of people; they are a serious attempt to define who lives inside and who lives outside the city. Metropolitan area definitions are not useful at all for counting city dwellers, since they are typically defined by county boundaries and are not a meaningful attempt to determine where the city actually begins and ends.  

What percentage of Texans are city dwellers? 75.35%. Although they comprise three-fourths of the population, the city dwellers occupy just 2.76% of the land. Over half of all Texans live in the urbanized areas of Houston, Dallas, Austin and San Antonio. (9.35% of Texans live in small towns and 15.3% in rural environments.)

Urbanized areas are in purple. The urban clusters are represented by the green dots.

Texas is the 15th most citified state by this metric, more urban than all the midwestern states except Illinois, and all southern states except Florida.  And it is more urban than the nation as a whole; nationwide, 71.4% of Americans live in one of the 486 urbanized areas.

Despite the fact that Texas is mostly a state of city dwellers, I'm sure we'll be bombarded soon enough by campaign ads from politicians posing by cow pastures and split rail fences. Because, you know, the real Texas is rural. 

This is how housing markets are supposed to work

Demand for housing has been rising in the Montrose/River Oaks area in Houston, especially from renters. Rents rose there at a 9.1% annual rate in January.

The difference between Houston and a lot of other cities is that it is still easy to add housing in Houston's nice, central city neighborhoods (unless your project has "Ashby" in the title). There are currently 15 apartment projects with 4,300 units under construction in the Montrose/River Oaks area. That's not "announced" units; that's 4,300 units under construction. For point of reference, only 3,089 building permits were issued for housing units of any type in the entire San Jose-Sunnyvale-Santa Clara metropolitan area in 2011.

Houston has a lot of needless land-use controls, including excessive minimum-space requirements and parking minimums,  but there really aren't many other places in the country where there is both strong demand for infill development and a regulatory environment that freely allows it. 

Comment moderation

I've been getting a lot of product review comment spam, and at an increasing rate. I will moderate comments for a while to see if that discourages it. Once the spam tapers off, I'll turn moderation back off. 

Give zoning an expiration date

I've been thinking about expiration dates for zoning.  I was prompted by Austin's draft comprehensive plan. There are things I like about the plan -- it advocates increased density on Austin's primary urban corridors, for example. But there’s also stuff to dislike. Such as this:

I’m sure most American municipal comprehensive plans contain one variation or another of the principle, “Protect existing neighborhood character.” Austin is no outlier here.

But this seems to me to be very bad policy for a long-range plan. Zoning’s fundamental problem is that it stifles change and adaptation over the long run; a long-range comprehensive plan should push back against that tendency rather codify it.

More from David Schleicher

Mark Bergen at Forbes has an interesting interview with George Mason law professor David Schleicher, whose paper, City Unplanning, I discussed a couple of weeks ago.  Schleicher's thinking hard about the political dynamics that lead large cities to systematically underzone. He correctly focuses attention on downzonings, which tend to occur quietly (if not stealthily):

1) Here’s the game theory part: Local legislators may prefer more development than we have now to less, but have stronger preferences for stopping development in their districts because these projects would hurt homeowners in their neighborhoods—either directly through things like increased traffic or indirectly through increasing the supply of housing, harming the value of existing houses.

This is a prisoner’s dilemma and absent a political party to organize the vote in local legislatures, one-by-one votes on projects will result in “defect” results, or situations where every legislator builds coalitions to block projects in their own district and nothing gets built. City Councils end up featuring what one journalist called the “ironclad principle of aldermanic privilege” or a norm that Members of the City Council have an absolute power to veto zoning changes in their district.

This makes a big city like a bunch of exclusive suburbs, at least for zoning purposes. Projects that would help the whole city, benefiting economic growth and reducing housing costs for newcomers or renters, lose out because the neighborhood in which they are located doesn’t want them and because each project is presented to the Council seriatim.

2) The process works differently for big developers and small ones. When a big new project is proposed, the politics look pretty ordinary, with a big developer fighting against local homeowners.

But big developers don’t care about “downzonings” or reductions in the size of the zoning envelope to current use, so that no one can build in a neighborhood as of right. The developers don’t care because they haven’t an investment yet, and the only concentrated and political powerful group are nearby homeowners who want to restrict development. These downzonings—which have become increasingly common—really matter because small developers can’t afford to go through the city planning process to get changes made.

Because we consider changes seriatim, small developers can’t piggyback on the lobbying strength of the big developers. Cities end up with a bunch of towers but strong limits on incremental housing growth—small new buildings, granny flats, and the like. This too pushes up housing costs.

That sounds a lot like Austin. Infill development in this town is conspicuous because the projects are so big. What you don't see much of here are small new multi-family buildings or granny flats. Note that downzonings don't have to be formal changes in zoning categories (i.e., from MF 3 to SF 3). The McMansion ordinance was a downzoning -- by drastically restricting the floor space and envelope of buildings, it made granny flats, garage apartments and duplexes less desirable in single-family neighborhoods.  The "superduplex" ordinance was also a downzoning, because it made duplexes more difficult, or at least less attractive, to build.  Austin should be moving in the other direction to permit small, multi-family buildings as of right in more places.

Forbes will publish the second half of the conversation tomorrow.

More fun with population density

Apparently, no matter how dense your city, it is at the cusp of becoming "too dense." Manhattanites feel this way; this New York Times article explores how many people Manhattan could hold. Manhattan's a pretty dense place, so it takes Kowloon Walled City to make the island look sparsely populated.  

Matt Yglesias makes the point that California would hold over 188 million people if it were settled at New Jersey's density. This would be a good thing, he argues, because California is a nice place to live and has a temperate climate.  There are problems with comparing states like this, though, because one state (such as California) might have a lot of inhospitable land.  

But there is no such problem in making comparisons between cities.  The NYT article made me wonder how large Austin would be if it were settled at the densities of some other American cities. 

According to the 2010 census, Austin had 790,390 people on 297.9 square miles, for a (standard) density of  2,653.2 people per square mile.  Obviously, Austin would be gigantic if it were populated at San Francisco's density of 17,170 ppsm (in which case Austin would have 4,327,000 residents) or at New York City's average density (in which case it would have 7,411,000 residents). 

But no one considers Dallas or Houston to be terribly dense places -- they are the prototypical suburban sunbelt megalopoli. Yet both are much denser than Austin. If Austin were populated at Dallas' density, it would have an extra 258,000 people. At Houston's density, it would have an extra 289,000 people.

A lot of the cities we think of our as our peers are much denser. If Austin were settled at Denver's density, it would have an extra 363,000 people.  At San Diego's density, an extra 402,000.  And at Portland's density, an extra 513,000. Columbus, Ohio is probably Austin's true "peer" city - it has almost the identical population, is home to a giant state university, and is the state capital. Austin would have to add 290,000 to reach Columbus' density.

 People around here would raise hell, of course, if the City Counil were to allow the kind of development needed to raise Austin's density to that of its peers. But the question is, Why? Obviously, some of the nicest places in the country are nice places despite significantly higher densities. It may even be the case that they are nice places because of significantly higher densities.  I've always thought it really weird for people to complain that rising density will ruin their quality of life when so many places with an equally good or better quality of life are already much denser.

PS. There's a cool 1989 German documentary on Kowloon Walled City on Youtube. Needless to say, life at a density of 3.2 million people per square mile was pretty strange.  H/t @MarketUrbanism.

Strong Towns

Last Wednesday, I attended a CNU-sponsored "curbside chat" by Chuck Marohn. Chuck is one of the founders of the Strong Towns movement.  He's a civil engineer so, naturally, he mostly talks about infrastructure. His basic message is that suburban sprawl seems cheap to towns and cities because the developers pay for most of the infrastructure. But that infrastructure has a finite lifespan. Eventually it must be replaced. And when the time comes to replace it, the tax revenue generated by the low-density, low-value sprawl pattern won't cover the cost.

Chuck's from a small town in Minnesota, so he has a different perspective on growth than those of us in Austin or other fast-growing places in the sunbelt.  A lot of Midwestern towns and cities stopped growing, or actually began to lose people, a long time ago. They are now saddled with worn-out roads and streets, leaky water pipes, and broken sewer systems that they cannot replace without outside help.  Some cities are deliberately abandoning some infrastructure, or even whole neighborhoods, because they can't cover the cost any more.

Cities like Austin should pay attention. Austin is a very young city. More than half the city was built within the last 40 years.  And it will continue to add new subdivisions a rapid clip, which will hold down the average age of our infrastrcture. But eventually the growth will end or slow to a trickle, and then we'll be stuck with a lot of aging infrastructure, and, thanks to our low-density growth patterns, a high ratio of infrastructure to people.

Obviously, cities should aim to grow their existing neighborhoods. Adding people to an existing neighborhood saves the need for new infastructure and provides more bodies to pay for the existing infrastructure when it wears out.

But my main reaction while listening to Chuck was, What do you think we're going to do about it?

Austin's comprehensive plan assumes the city will add 750,000 people over the next 30 years. I think that's optimistic -- 350,000 to 400,000 is more likely, based on last decade's rate of growth. The metro area will likely add well over a million, though.

Almost all of that growth will be greenfield. There is no political will to allow the kind of upzoning that would be necessary to accommodate even a small fraction of that growth within existing neighborhoods. Consider central Austin. It added just 8,000 people between 2000 and 2010, while Austin proper added 133,000 and the metropolitan area added 450,000 people. The central city did a bit better job adding structures than people -- its housing stock expanded by 15,000 units, or roughly 12%. But the last decade was the decade of condo towers sprouting downtown, the densification of West Campus, the Triangle, Mueller, condos in East Austin, and massive greenfield development in southeast Austin. And all that development lead to a piddly net increase of 12% of the housing stock. 

Don't get me wrong. I think we should let central Austin get much denser, for many reasons. Austin is lucky because people actually want to move to the central city. Lots of cities spend a lot of time and money trying to solve the problems caused by central-city flight. But given the contentiousness of last decade's development battles, I don't see Austin allowing the kind of development we need to house even 50,000 new people in the central city.