Austin Contrarian

The coming robocar apocalypse

How will driverless cars affect the built environment? A lot of the discussion lately has focused on whether driverless cars will affect the demand for rail. Michael Lind thinks they will make rail obsolete. Alex Block, sensibly noting that a change in technology won't change the relevant geometries, disagrees.

Matt Yglesias (responding to a Randal O'Toole post that has been taken down) foresees a parking bubble:

[R]ight now every metropolitan area in the United States contains many, many more parking spaces than automobiles. When you're at work, the space allocated for your vehicle at home sits there empty. When you're at home, the space allocated for your vehicle at the office sits empty. Malls build parking to accommodate demand during peak hours, and the spaces mostly sit empty off-peak. But if the cars could drive around without a human pilot, there'd be no need for such lavish supplies of vehicle storage. In principle, a metro area could get by with fewer than one parking space per car since even at minimum-demand times a nonzero quantity of vehicles would be in use. That's probably extreme, but right now depending on how you count we have somewhere between three and eight parking spaces per car. If the cars don't need to sit idly waiting for you until you want to leave (imagine a world of cheap, ubiquitous taxis) that number is going to become totally ridiculous. After exploding for about 60 years, the torrent of parking construction is going to halt very suddenly and then start shifting into reverse. That should even make some rail lines more useful. Commuter rail stations, for example, will no longer need to choose between park-and-ride and transit-oriented development models. Every station will be a little TOD neighborhood, and people from further away will get dropped off and picked up at the station without needing to worry about storing a car there.

I agree with Matt that if robocars become ubiquitous, then we will see a reduction in demand for parking. But we will not see a reduction in demand for vehicle storage.  Vehicles will continue to exist, to occupy space, whether someone is behind the wheel or not. Driverless cars will simply encourage a shift from static vehicle storage (a/k/a "parking lots") to dynamic vehicle storage (a/k/a "streets").

Donald Shoup has harped (pdf) on the phenomenon of "cruising for parking." It is one of the hidden costs of underpriced street parking. Underpricing street parking causes every spot to be filled, which, in turn, causes drivers to circle the block, searching for an open spot or waiting for one to open up. Shoup surveyed a number of studies which found that, on average, 30 percent of the traffic flow consisted of drivers cruising for parking.  Underpriced parking, counterintitively, means congested streets.

There is a built-in check on cruising for parking, though: every driver eventually wants to stop and get out of the car. Drivers aren't cruising for fun; they are cruising in the sunny hope of finding a place to store their vehicles and get on with their business. If a particular part of town gets a bad parking rep, then a lot of drivers won't even bother going there. The fact that, today, drivers must tether themselves to their vehicles while in motion limits the amount of congestion from cruising.

With robocars, though, we'll witness a new phenomenon: cruising in lieu of parking. Rather than cruise Congress Avenue for fifteen minutes looking for a spot so I can get a taco at Chipotle's, I'll simply have my car drop me off at the door and command it to spend the next 30 minutes getting in everyone else's way while I eat. It's a pretty simple cost-benefit calculation: add the cost of parking to the cost of the anticipated time it will take to find a parking spot; if that cost is higher than the cost of gasoline, then just order the car to cruise.

It won't take a lot of people engaging in this practice to congest the streets. Just imagine, say, an extra 200 cars continuously cruising up and down Congress during the middle of the day. What's worst is that the people who will benefit from the practice will incur none of the costs. On the contrary, the more gridlocked the traffic, the less gasoline their cruising cars will burn.

In fact, this perfectly rational practice will probably be so harmful, so patently selfish, so despised that it will be necessary to outlaw it. Which means everyone will still have to find a spot for his car, driverless or not. Which means that, despite the title of this post, we might not see a robocar apocalypse after all, or a parking bubble, either (other than the existing bubble that local governments have created with underpriced street parking and mandatory parking minimums.)

Does East Austin really have a high level of homeowner turnover?

These statistics are not meaningful without a baseline to tell us what is "normal":

More than 1 in 3 people who bought homes in some of East Austin's oldest neighborhoods in 1992 or earlier moved out of those homes after 2002, and close to 1 in 3 of remaining homeowners are delinquent on their property taxes, according to a new study calling for community solutions to help people stay in their homes.

Released Tuesday by two nonprofit groups, the study concludes that rising property taxes in those East Austin neighborhoods have squeezed out many low- to moderate-income families. Though the report calls the turnover of homeowners alarming, it says that hundreds more longtime residents still reside in East Austin and that helping families stay in those homes is better than traditional approaches of building new affordable housing.

"This (report) gets to our whole concept of the American dream, and what does that mean, and what should we expect?" Raul Alvarez, one of the report's authors, said, adding: "In this day and age, we need to be able to purchase our home and be able to live there all of our lives and die there if you choose. That dream has changed dramatically for some of these eastside residents."

A former Austin City Council member, Alvarez is president of the East Austin Conservancy, which issued the report with People Organized in Defense of Earth and her Resources (PODER).

"We talk about sustainability in Austin. I think part of the sustainability is keeping people in the neighborhood," Susana Almanza with PODER said. "What we don't want Austin to become is an urban core only for the rich."

The report, which studied Chestnut, East Cesar Chavez, Holly and parts of Rosewood neighborhoods, does not address how homeownership turnover and tax delinquencies in those neighborhoods compare with other areas of the city, nor where those longtime homeowners moved.

It is based on findings gleaned from Travis Central Appraisal District records.

Karen Paup, a housing advocate who works on affordability issues, called the turnover rates in the report striking.

"There's probably not another neighborhood (planning) area in Austin where we would find that 33 percent of the houses ... had that high of a turnover in a 10-year period," said Paup, co-director of the Texas Low Income Housing Information Service, a nonprofit research and advocacy group.

I don't have  a copy of the report and would appreciate someone forwarding it to me. But these turnover figures aren't striking to me.

"We need to be able to purchase our home and be able to live there all of our lives and die there if you choose." Very few people do this. It is not the norm. Most people move several times, even after reaching adulthood. According the American Community Survey's 2010 5-year estimates,  56% of Austinites living in owner-occupied homes in 2010 moved into their homes between 2000 and 2010. Only 18% of Austin's homeowners have lived in their homes since 1990. (Renters are even more mobile: 94% of all renters moved into their homes between 2000 and 2010.)

I haven't done a systematic study of other Austin neighborhoods, but these East Austin neighborhood appear to be more stable than a couple of other old, central Austin neighborhoods for which I already had data. Here's the ACS data for Tracts 9.02 and 10, which roughly match the "study" boundaries above, and Tracts 13.03 (part of Zilker) and 13.05 (Bouldin Creek):

The ACS numbers are just estimates with pretty wide margins of error.  Still, they suggest that the two East Austin tracts have at least as many long-time homeowners than either of the two tracts just south of Lady Bird Lake.  ACS estimated that 43% of the homeowners in Tract 10 (as of 2010) moved into their homes before 1990. It estimated that 35% of the homeowners in 9.02 moved into their homes before 1990. The comparable stats for Tracts 13.05 and 13.03 are 32% and 22%, respectively.

The City-wide average, again, is only 18%, although this is of course skewed by new construction and not as meaningful as a comparison with neighborhoods of similar age. But no one has done such a comparison. Without it, we can't tell whether East Austin is experiencing an unusual degree of turnover. And without some evidence that East Austin is experiencing an unusual degree of turnover, the advocates haven't shown that property taxes or other economic hardship is driving people out of East Austin.

What is the real constraint here?

Austin badly needs more downtown office space. Austin's downtown Class A rents are higher than either Dallas or Houston's, which is absurd. But it is difficult for developers to line up money to build an office building on spec, particularly without a large anchor tenant to soak up a few floors. Austin doesn't have many firms looking for large blocks of downtown space.

Cousins Properties will try, though. It has announced plans for a 30-story office tower at Third and Colorado streets. The tower would have  about 390,000 square feet of office space and 6,000 square feet of street-level retail. Twelve of the 30 stories would be for a parking garage with 900 spaces. It would be downtown Austin's first new high-rise office building since Frost Bank Tower opened eight years ago.

Given the intense demand for new office space downtown, why limit the project to "just" 30 stories? On net, this building will yield just 17 or so floors of office space. Why not more?

Downtown zoning regulations limit the building to a floor-to-area ratio ("FAR") of 8:1. The owner got approval in 2009 for a FAR of 12:1, but the zoning ordinance restricted the increase to a hotel use, and the developer needs 15:1 FAR anyway.  Althought it's ridiculuous that central business district zoning limits this downtown site to around eight floors of office space (parking doesn't count against FAR), that's not the real constraint on this building's size. The developer is asking for 15:1 and I expect it will get it.

I suspect the real constraint here is parking. Drivers will only drive up so many ramps in order to park their cars. The taller the garage, the bigger the pain it is to park on the top floors. This puts a practical cap on the number of floors a parking garage can contain. The developer is proposing a 12-story garage, which is an aggressive estimate of that cap. It's safe to say that this plan would maximize the amount of parking on site.

Despite all that parking, the plan is proposing "only" 2.3 parkings spots per 1,000 sf of office space -- less than that, really, since some of that parking should be allocated to the ground-floor uses. Although this is  more parking than the city requires, it is less than conventional wisdom says new office buildings in downtown Austin need, which I've been told is around 3 spaces per 1,000 square feet.

I'm not questioning the feasibility of this plan. As office rents rise, the opportunity cost of a parking spot increases, which should lead to less parking per square foot of office space. And I'd be delighted to see the ratio of parking to office space decline in downtown Austin. But the extreme height of the garage combined with the aggressive parking-to-office-space ratio suggests that the developer has proposed 17 floors of office space because it can't fit in enough parking for 18 floors. Put differently, the developer is building as much leasable space as its maxed-out parking garage will support.

Run urban rail up Guadalupe/Lamar

Jeff Wood at The Overhead Wire has been running a series on the history and politics of Austin's urban rail endeavors. Today he's posted an immaculate argument for running the line up Guadalupe and Lamar rather than out to Mueller.  The aerial photo below, which contrasts the proposed line (orange) and a Lavaca/Guadalupe/Lamar alignment using current bus ridership, makes the case succinctly:

Guadalupe and Lamar are at capacity and have been at capacity for decades.  Planners are shying away from them, evidently because they are too congested. But "being too congested" is the reason to make this massive investment in rail in the first place. Reserved guideway transit is the only way to significantly increase the capacity of these prime gateways to UT, the Capitol Complex and downtown.   Spending a couple hundred millioin on the four-mile segment from UT to Mueller is largely a waste at this point -- that's not a congested route, and it doesn't have much existing transit ridership.

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.