Skip to content
June 23, 2014 / Dan Keshet

Playing around with a gridded transit system

Yesterday, I was exposed to a new tool that came out of Code For America: TransitMix.  It’s intended as a tool for allowing lay people (and eventually professionals) to draw up transit systems.   The map that I drew up (ideas you like are from Brad Absalom; ideas you dislike are from me)is a very preliminary interpretation of how ideas gleaned from reading posts on Jarrett Walker’s blog about a grid-based, transfer-happy system could work in Austin.  For this post, I will call the system I drew up Reimagined in the spirit of Houston’s System Reimagining (on which Walker consulted). For speed, I didn’t take on MetroExpress (i.e. 983, 985, 990), the Flyers (e.g. 103, 142, 171), or the UT Shuttles (e.g. FA, WC, LA).  I also ignored off-peak times for the same reason.

I constrained myself to reality in a few major ways: 1) using the same number of peak-hour buses that I calculated Cap Metro to use within MetroBus / MetroRapid (136 buses).  This gives a rough idea that the new system has similar total expected costs.  2) I attempted to keep coverage for as many locations currently served by a bus as I could, 3) I did my best to give average run speeds based on real Capital Metro runtimes along segments, including rest times for driver breaks.

I only drew this up for Devil’s Advocate purposes at this time–I’m intrigued by it enough to throw a Saturday afternoon into making it, but I’m sure there are issues I haven’t thought about.  It’s a very first cut, just an attempt to follow the principles and see where it leads.

Read more…

June 8, 2014 / Dan Keshet

Should we ban part-time taxi drivers?

Lyft and Uber have (re-)arrived in Austin, offering free rides for a couple of weeks, despite the City of Austin declaring they are operating against the law.  You can’t beat free, so I’ve been taking a lot of these rides.  All of my drivers have been part-timers, looking to supplement income.  This is in stark contrast to taxicab drivers, who are pretty much all working more-than-full-time hours.

The economic reasons why taxi drivers are full time make a lot of sense. Cabs are dedicated-purpose vehicles, going for the most part unused when they aren’t working.  Taxi licenses are an extremely scarce resource due to city regulations.   The taxi industry will find a structure to make sure these two expensive resources are working as often as possible.  Taxi licenses are issued to individuals, though; so this means that any structure that ensures that we have a lot of usage of our limited taxi licenses also ensures that our taxi drivers are working overtime.  The form this takes in Austin is high costs for either purchasing or leasing a taxi and high “terminal fees” for getting the right to use the taxi companies’ dispatch system.  Every month, each taxi driver starts deep in the hole and has to work a lot of hours just to break even, to pay for the fixed costs of their cab and their terminal fee.

So how do Uber and Lyft turn this dynamic around and allow for part-time drivers?  Uber and Lyft (whatever the legalities of their operations) are not limiting themselves to a city-capped pool of taxi licenses, nor are they limiting themselves to cars that are used exclusively as taxis. If a driver decides not to work Uber for an afternoon, a week, or a month, this doesn’t take the expensive fixed cost of a taxi out of their network, nor does it take the scarce resource of a taxi license out of their network.  Because of this, Uber and Lyft can do away with the terminal fee and instead just take a percentage of each ride as a dispatch fee.  Felix Salmon does the math and finds that, if the numbers are to be trusted, a full-time driver with Uber makes pretty good money compared to most taxi drivers, but a part-time driver does very well indeed.

There are many, many issues at play with Uber and Lyft, but from a very zoomed-out perspective, current city regulations make it uneconomical to have part-time drivers, and Uber and Lyft have come up with a model to change that.  Now, why we do we care?  Because taxis have very variable demand.  At peak hours, the number of licenses Austin hands out aren’t nearly enough to keep up with the demand.  But because drivers are starting off so far in the hole, they can’t just work peak hours; they have to work additional hours even if they’re very poorly paid for those hours, because they have to pay off the high fixed costs. So our current structure works for neither drivers nor riders: drivers have to scramble for every last ride in off-peak, but there still aren’t enough of them to give rides on-peak.  A structure (whether Uber/Lyft or something else) that allows drivers to come online during peak but doesn’t force them to work tons of off-peak hours could be beneficial to drivers and riders alike.  This presents a test of sort for any change in regulations.  Will it allow the economics to work for part-time drivers?  If not, then it really hasn’t accomplished much.

May 27, 2014 / Dan Keshet

A new slogan: abundant housing

AURA, the urbanist group in Austin I’ve been so excited about, has adopted the language of “abundant housing.” The concept is simple: build enough housing for everybody who needs a home. Although abundance and affordability are intimately linked, the call for enough homes for everyone doesn’t require invoking affordability. It’s a strong statement of inclusivity and social justice on its own.

Abundant housing makes all housing more affordable

Even if you don’t buy my argument (and overwhelming consensus among economists, real estate people, etc.) that abundant housing leads to affordable housing, you should stand up for abundant housing because everybody deserves a home.  But I’ll flesh out the link anyway. I’ve compared the housing market to a game of musical chairs. If you don’t have enough chairs to go around, the competition is going to be intense not just for the last few chairs, but for every chair. Similarly, if there aren’t enough homes to go around, the competition for homes will be intense up and down the market. This is the situation today in Austin and in cities throughout the country: rising prices, cash-only sales, homes selling sight-unseen. The competition for homes is intense at all levels; even those who can definitely afford some housing are forced to contend with rising rents. Abundant housing aims both to make sure those at the bottom have a place to live and everybody else isn’t subjected to the intense competition that makes our current housing markets so hard.

Targeted affordability isn’t enough

Affordable housing can refer to mandated Affordability or housing that’s just affordable. Whether the specifics of the affordable housing strategy are bond-driven Affordable Housing, Affordable Housing setasides in new development, or targeted relaxation of laws that only allow new housing that will likely be affordable (e.g. very small units), these strategies are all based around the idea that the best way to improve affordability is to create new housing that will enter the market at the lowest end. In its worst form, you see not only a preference for housing targeted at the cheapest end of the housing market, but opposition to the creation of net new housing because that new housing is at the higher end.

But in a game of musical chairs, the intense competition for chairs doesn’t come from the fact that there are too many thrones and not enough folding chairs; it comes from the fact that there aren’t enough chairs, period. If there are ten fewer chairs than there are contestants, whether you add ten thrones or ten folding chairs, everybody will now have a seat. In this view, discouraging people from adding seats because they aren’t targeted at the particular folks who lost last round is perverse; you are hurting those folks’ chances of finding a seat, not helping them. Abundant housing is built around the same idea: policy that allows new housing to be created easily will allow more homes to enter the market, enough that everybody can find a place to live. With the competition between buyers/renters being less intense, rents won’t rise as much across the market. Whether that’s done through small new units cheap enough to hit the middle-to-bottom of the market, bond-supported Affordable Housing units, or new luxury condos absorbing the demand from the richest, there will be housing for all.

Lack of abundant housing overwhelms Affordable Housing

Affordable Housing policy is necessary to help those who need it, even when there’s enough housing to go around. For the musical chairs analogy, in a game with as many chairs as contestants, an adult might find themselves next to vacant chairs designed for children; a contestant with a mobility impairment might not be able to reach a vacant chair because the path is blocked. This is analogous to a well-functioning housing market in which there are still some folks who need a hand, perhaps temporarily because some bad thing happened or they simply lack funds; perhaps permanently if they have a reason they will never be able to make rent. But these intervention strategies are overwhelmed when there are simply too few chairs to accommodate all contestants. If there are fewer chairs than contestants, no amount of targeted interventions will help. They merely result in a different set of folks going without a chair. But fixing the abundance problem, the Affordable Housing efforts can go back to serving specifically those who need it even when there are enough chairs to go around.

May 22, 2014 / Dan Keshet

Another positive development (accessory units)

Following up on the positive development of Chris Riley and Bill Spelman pushing to lower parking and density requirements for small units along dense corridors, Riley is back at it, this time teaming up with Mike Martinez to lower parking and driveway requirements for accessory dwelling units (secondary structures in the backyard of single-family homes).  Chris Bradford and Steven Yarak have great rundowns on why the ordinance is needed so badly and how it could be even better if it applied to more ADUs.

Chris and Steven can give you a better overview than I can of why ADUs are needed and why parking and driveway requirements make it so hard to build.  So I’ll add a perspective that is so obvious it shouldn’t need to be stated, but now and again we do well to repeat: parking and driveway requirements are unfair, backwards, and detrimental all around.

I live in Austin without a car.  At many apartments I’ve lived in, a solid chunk of the land was (un)used for the worst possible thing: an empty parking space, reserved for the car I don’t have.  That land could’ve been a garden absorbing rainfall and preventing runoff; it could’ve been an outdoor patio, providing me and the other residents with a place to enjoy the weather; it could’ve been more housing, allowing for me to split the costs of housing with another resident or roommate.  Is it useful or fair to require every resident of Austin that doesn’t have a car to still pay the costs of maintaining an empty parking space (and often driveway) next to their home?  Is it beneficial to the city or the planet to encourage residents to drive, over using choices like riding public transit or bikes?  (Crib sheet: answers are no and no.)

Even if I did have a car, though, requiring a reserved parking space for my home still wouldn’t make sense.  I lived in each case on streets with a significant share of their street space reserved for the sole purpose of parking automobiles. The city goes to almost incalculable expense to reserve space along almost every street for no purpose other than parking cars; yet frequently that space goes unused.  It is as if the city spent billions of dollars establishing swimming pools throughout every neighborhood, only to mandate that most private homes have swimming pools to ensure that the public swimming pools didn’t get overused.  Except, subsidizing swimming to that extent, while not a great use of funds, might at least be good for the health and enjoyment of the city, while subsidizing parking encourages emissions and traffic.

Likely, some of those who oppose the reduction in parking and driveway requirements don’t really want the parking requirements specifically; they just know that without them, there would be more ADUs, and they oppose that.  To the extent that this is true, it is a real shame.  The city requires a ton more parking spaces, driving up costs, harming the environment, and subsidizing traffic, all because some who couldn’t get the political forces necessary to achieve their (bad) objective of banning ADUs found a way of doing it by making it practically banned instead of actually banned.

I’m far from the only, first, or best person to make these points about parking.  If you’re interested in learning more, I suggest checking out Graphing Parking, a blog by Austin’s own Seth Goodman; Reinventing Parking, the internationally-themed group blog he participates in, and the canonical work, The High Cost of Free Parking, by UCLA professor Donald Shoup.

May 14, 2014 / Dan Keshet

My email regarding Uber / Lyft / Sidecar, etc.

Mayor, Mayor Pro Tem, and Council Members–

I urge you to vote yes on resolutions 24, 25, and 26 regarding transportation networking companies. For people like me who do not drive, paid rides are an essential emergency backup. However, I have had so many bad experiences with taxi companies telling me they do not have enough capacity to give me a timely ride that I cannot consider them a reliable backup.

I use taxis when I am too tired, too sick, or carrying too much to walk or use public transportation. I have used taxis to take friends home from the hospital, to take pets to the vet (not allowed on Cap Metro buses), and to take myself to doctors’ appointments that weren’t on transit routes. I have been told that I will have a long wait because most drivers don’t want to work off-peak and that I will have a long wait because it was peak hours and most drivers were already occupied. Waiting for a long time for a ride to a hospital is the sort of defining experience that makes many people give up on living without a car.

Living in Austin without a car is difficult enough. Making it easier to sell rides will give more security to those of us who cannot drive to know that when we really need it, there will be a timely ride for us.

May 11, 2014 / Dan Keshet

Let’s not make a deal

If you hang around the Austin (and presumably other city’s) zoning codes long enough, you get used to deals.  I’m not talking about compromises, where one side wants a height limit of 40′, the other a height limit of 100′, and they settle for 60′.  I’m talking about requirements built into code that a landowner can get out of if they satisfy some other requirement.  This principle is built so deeply into code, in fact, that it’s almost impossible to avoid.

The Vertical Mixed Use (VMU) zoning overlay allows a landowner to choose between the “base zoning” of a district or VMU zoning.  VMU zoning significantly reduces some limitations (FAR, density, see the linked Austin Contrarian post for definitions and details), and replaces them with different requirements: detailed specifications on acceptable building designs and requirements for some units to be designated as Affordable Housing (see link for definition of Affordable Housing).  Similarly, the University Neighborhood Overlay (UNO) allows landowners to build significantly more on a single property than they would be allowed under base zoning, but only if they agree to an extraordinarily detailed specification for how their building might look like.  (I continue to be amazed that these specifications forbid developers from developing new buildings that look historical because this “results in the devaluation of the real thing.”)  Transit-Oriented Development overlays work similarly; allowing greater densities in certain locations, as long as you follow certain design specifications.  The Downtown Density Bonus allows greater heights in exchange for funding Affordable Housing.  At a smaller scale, there are deals that allow developers to opt out of minimum parking regulations in exchange for providing bike lockers or providing parking for shared car services (e.g. Car2Go, Zipcar).

In general, I find myself in favor of each of these deals. VMU, UNO, and the DDB all provide an opportunity for a development that’s more of a proper scale for the locations of the building along major thoroughfares, or near UT or downtown. I don’t always think what the buildings are required to do in exchange is for the best, but it’s worth it to have that option.  Similarly, I’m in favor of vastly lower minimum parking regulations; if buildings provide bike lockers or Car2Go spaces, more to the better.

But for the short-term upside, there’s also a long-term downside: if either side of the deal changes, the whole deal will need to be renegotiated, making it much harder to make any changes.  When Council Members Chris Riley, Bill Spelman, and Mayor Pro Tem Sheryl Cole introduced an ordinance to explore the possibility of lowering minimum parking and density requirements for microunits along VMU corridors, the major objection expressed at Council was not toward the idea of allowing greater density or less parking itself, but rather that it would gut the existing VMU deal of less parking in exchange for more Affordable Housing.  (Again, see link for definition of Affordable Housing.)  Lowering minimum parking regulations citywide would have an effect on existing deals involving VMU, TOD, bike lockers, car2go, and thousands of ad hoc deals in which landowners agree to waive some development rights in exchange for reduced parking requirements.  It would be a nightmarish negotiation.  Each deal creates an opportunity for lowering the minimum parking needed in one situation, but also creates a constituency opposed to a broad-based reduction.

I’m not sure how to handle this situation–most new deals that come up are tempting, as they offer greater flexibility to build something of a more appropriate density than base zoning allows.  Some of them (VMU, UNO, DDB) have positively transformed the city, allowing far more people to find homes.  Yet, I fear that the more layers of deals that get added, the more difficult it will be to unravel. Small decisions like lowering minimum parking regulations even 10% will be impossible to make because of the number of stakeholders involved in the negotiations.

Fortunately, I feel very excited that this was one of the primary diagnoses CodeNext made [large PDF] in their review of Austin’s current Land Development Code: base zoning districts are ineffective, leading to complicated layers of opt-in, opt-out regulations.  I hope that a good solution can come up with to unravel these deals and that any future deals are made with caution.  A well-meaning deal that advances two priorities in the short-term can lead to paralysis in the long-term.

May 8, 2014 / Dan Keshet

In light of Hancock tunnel, whole northern segment of rail needs rethink

I was appalled by the sticker price for a tunnel to get rail part of the way north from the Hancock Center to the Highland Mall.  This sticker price has come down: from a range of $230-290M to $220M.  But it is still truly appalling: the capital costs for that short tunnel alone would be about twice the capital cost of the entire MetroRail system which is forcing a tunnel in the first place and more than 4 times the capital cost of the entire MetroRapid bus improvements that led planners to rule out a route west of the University. It would cost more than going from Travis Heights all the way to 15th Street, including a brand new signature rail/bike/pedestrian bridge over Lady Bird Lake and tracks through dense downtown.

I wasn’t alone in being appalled. At a Transit Forum hosted by Austin Monitor and KUT, I asked a question of Council Member (and CapMetro board member) Mike Martinez on the Hancock tunnel via twitter, and he said that if we had to tunnel (as Project Connect says we must), the costs go up “exponentially.” Mayor Leffingwell has said in the press that he’d like to see the $1.4B price tag for the whole route be brought down to $1B. The Hancock tunnel is an obvious first place to cut. Many early supporters of Project Connect’s plans are now cautioning against including a Hancock tunnel on the ballot. And in project lead Kyle Keahey’s testimony, he sounded almost desperate to find a way to avoid the costs of a Hancock tunnel.  Even if decision makers hold their noses and vote to advance the entire $1.4B project, there’s little reason to believe that voters will do the same in November.

In light of this setback, we have to rethink more than just the tunnel, but rather the entire northern segment of the urban rail. Originally, Highland was sold as a low-cost segment with room for growth in the Highland Mall redevelopment, the old Concordia site, and along the newly re-zoned Airport Blvd Transit Corridor. Whatever you thought about the potential for these destinations to encourage ridership on a rail line (and I was skeptical), rail reaches none of them without the tunnel. Instead, it would merely run along a street not designated as a Transit corridor, through a neighborhood of mostly single-family homes that have largely struggled against becoming another University-adjacent density hotspot like West Campus, and then come to an ignominious end at a golf course and an auto-oriented strip mall.  Without the Hancock tunnel, the whole raison d’etre of this route goes away.

If that were all, that would be disappointing. A medium-cost, low-ridership train is not great.  But this low-ridership segment of the route threatens far worse: it could make future rail plans with higher opportunity impossible. This has happened to Austin twice already.  The low-ridership, low-cost MetroRail ended up badly hemming in this route, forcing a tunnel that cost twice as much as the system itself, something that few could’ve foreseen when MetroRail was being approved.  The low-cost MetroRapid gave planners reason to fear placing this rail route along the city’s most productive transit corridor.  A low-ridership segment from UT to Hancock could do immense damage to future flexibility in laying rail lines.  Funding a rail line that hits a hard stop before it reaches its density could end up preventing funding for parallel rail lines along more dense parts of Austin as “duplicative.”  And for what?  Unless city planners have a secret plan to get rid of MetroRail, we will face the same high tunneling costs next time we look into extending rail.  We will have essentially backed ourselves into a dead-end that will be very difficult and expensive to get out of.  Even for those who believe that rail should go to Mueller, this would remove routing flexibility from any planning efforts.

The best thing for CCAG and City Council to do is say that, in light of the new information they have on the costs of the Hancock tunnel, they will forego a vote on the more controversial, risky, poorer-thought-out northern segment of the route, advance the southern segment on its own, and come back to revisit the question of the northern route later.


Get every new post delivered to your Inbox.