The Austin City Council is poised to make a decision on urban rail’s route based on incomplete data. At the heart of this predicament is a 2010 document entitled the “Central Austin Transit Study” (CATS). This document makes a set of fundamental assumptions about transit politics born out of the defeat of the 2000 rail referendum. Much has changed in the last thirteen years and even more will change by the best-case-scenario-ribbon-cutting for rail in 2021. Council needs a new, richer set of data to make the best decision for their Austin constituents.
CATS focuses on evaluating the stops on a route (‘alignment’ in transit jargon) and the vehicle type (‘mode’) for thirteen connections in a pre-determined area that includes the University of Texas, the Mueller redevelopment area, the Central Business District, the Capitol Complex, East Riverside, and Bergstrom airport. The thirteen connections are visualized in the image below; darker color indicates the CATS recommended route.
CATS does not explain why this is the optimal initial sequence for urban rail (the vehicle technology it finds most attractive). It certainly demonstrates positive benefits, but there are no comparisons with alternative routes outside of the thirteen connections.
The study provides two ridership estimates with the highest being 27,600. I could not find source code implementing the algorithm or the weights of factors described in the report narrative or the data file used. CATS does reference past studies and their associated public feedback process and calls the 2006 “Future Connections” study the ‘source study’ for its rail alternatives evaluation.
So, perhaps the “Future Connections” study (FCS) explains why the CATS recommended alignment is optimal? No, it does not. Instead – just like its child – FCS focuses on a constrained area and is fairly direct about its view of Central Austin transit as intended to ‘support the planned Capital MetroRail system and serve major destinations in Central Austin not directly connected to the rail system.’ Right below that statement in FCS is a map of the study’s focus area and a few pages later there is a list of twelve familiar-looking connections the study evaluates.
FCS goes and scores the narrow set of connections according to multi-criteria rubric (ridership, land use, cost, etc.) to craft the best route for MetroRail-supporting circulator. You can notice bands of green delineating suggested routes for rapid buses – large, technology enhanced buses that can communicate with signals. Finally, it is important to note that FCS scores a streetcar vehicle technology ahead of buses. It did not evaluate urban rail.
At this point, we are still wondering why a circulator supporting suburban commuters is the optimal first sequence for urban rail in the Austin we anticipate for 2021. We know that streetcars and urban rail are better than bus and that a circulator supporting the Red line has some ridership benefits, but we don’t know why it is the best choice. Better is not the same as optimal. And why are high ridership routes being served by the rapid bus technology? To figure this out, we have to consult the 2004 plan that spawned FCS: the “All Systems Go! Long Range Transit Plan” (ASG).
ASG was the first major public input process and study undertaken by CapMetro after the failed 2000 urban light rail referendum. The 2000 referendum proposal submitted to the Federal Transportation Authority featured a 20-mile light rail transit system with 26 stations running north-south from McNeil Road to Ben White Boulevard, and east-west from the central business district to 5th and Pleasant Valley. It estimated a ridership of approximately 37,400 weekday boarding by 2025. I was not able to gain access to a document detailing the methodology for those projections or the source file for their calculation. The proposal’s route can be found in the image below.
ASG replaced the failed referendum route with rapid bus. The remaining ASG documents do not provide a data-driven reason for this substitution. Instead, the document justifies its direction by noting the thousands of individuals that provided public input.
Obviously, the outcome of the referendum weighted on the minds of the facilitators and participants of ASG, which was used to advocate for the successful Red Line referendum. Thirteen years later, it appears our transit policy is still being driven by the child and grandchild of ASG.
When I walked into a Project Connect open house I was expecting fresh data-driven comparisons that yielded optimal recommendations. Instead, I was just offered recommendations based on CATS and its ancestors. And no data or model, just visuals. An example chart from the open house follows.
I didn’t feel empowered to make a decision on whether these recommendation were best. One could complain if there was an instinctive feeling that something was absolutely terrible and awful. But without comparison data, how would an empiricist like myself even know that?
It turns out that the influential members of Capital Area Metropolitan Planning Organization’s Transit Working Group (TWG) also were curious about the suburban commuter circulator route recommended by Project Connect. Nowadays, the alignment is typically referenced by the much warmer feeling-inducing moniker of ‘the Mueller route’.
In addressing their concerns of why a high ridership corridor like Guadalupe/Lamar was not selected, the City of Austin’s Transportation Director made two arguments: (1) that the Hyde Park to UT route was not highly-scored within the CATS study and that (2) there was already a federally-funded rapid bus in the corridor. The relevant part starts at 7:59 in this video.
The Hyde Park to UT argument is tangential. The Director was referencing a potential connection in the CATS circulator as it was scored by the subjective formula of the study – it has little to do with overall ridership of Guadalupe/Lamar. And the implicit argument that there is not enough value is directly contradicted by his second point, which is that the route is so valuable that it is where rapid bus is being deployed to first (!).
While his argument might be compelling if Austin was deploying urban rail in the same corridor a year or two after the 2014 start of rapid bus, the Director pointed out at the May 11th 2012 TWG meeting that a ribbon-cutting for rail is at best six years away if we get funding from FTA immediately and the project progresses quickly. By then, the rapid buses would have logged countless miles; and in any event, they could be re-purposed to a different corridor. Finally, the local cost of the rapid buses is $9 million. Is it worth it to pick a potentially sub-optimal urban rail route costing hundreds of millions of dollars because of such a relatively small investment?
Just to reiterate the obvious: there was no defense of the Project Connect recommendation on the basis of recent ridership comparisons, just assertions about ‘optics’ that FTA might not ‘like’. Other Mueller route supporters will similarly offer justifications based on ‘development impact’ or likelihood of Tax Increment Financing – but they are not backed by detailed or comparative data. It’s definitely possible that Mueller is the optimal first sequence, but the comparative evidence is not in the public domain.
One piece of data we know with certainty is who funds transit in our region. While CapMetro has a regional focus, it is funded by the sales taxes and bus fares paid in Austin. Sales taxes in the CapMetro’s service area are its primary source of revenue according to CapMetro’s 2013 budget.
And of that revenue, almost all of it is generated by sales taxes emanating from Austin. Certainly, some residents of Leander buy their lunch in Austin and so on. But there’s no way to argue that population centers outside of the Austin city limits are meaningful financial contributors. Austinites pick up the tab for the externalities of suburbanites choosing to live far, far away from their places of employment to get additional residential square footage. The data below is from the CapMetro Comprehensive Annual Financial Report for FY 2011.
The clarity of who pays for transit juxtaposed with the incomplete set of data about the optimal initial urban rail sequence should catalyze action by Council to seek out richer data on alternative rail routes that addresses the weaknesses in how the narrowly-tailored CATS is being used as the justification for Mueller as the initial urban rail sequence. CATS is about the best allocation given the connections available to a central circulator – it is not justification for Mueller as the first urban rail project. In the spirit of keeping things weird, the Council could call the new analysis the Deciding Optimal aligGment Study. I certainly hope the data is both loud and has some bite.
It’s not just some egghead blogger pointing this out. Two externally-powered reviews of our urban rail organizing also concluded that a more transparent process with better data is needed. The hot-of-the-presses Urban Land Institute Rose Fellowship report indicated that we should urge the public input process to: “Show your work. Reinforce unprecedented openness and transparency.” Additionally, this January’s APTA peer review observed that: “The Project Connect and Urban Rail project plans are lacking complete information regarding projected ridership. Ridership projections are particularly important for the further development of Urban Rail.”
If you are interested in helping bring about a transparent and data-driven decision on the urban rail route, then you might want to check out Austinites for Urban Rail Action (AURA), a new group formed by a small band of committed Austin citizens.
Thanks for the analysis and keep it coming. This is a conversation we need to have. I cringe a bit when I read this “Austinites pick up the tab for the externalities of suburbanites choosing to live far, far away from their places of employment to get additional residential square footage.” How does the lack of an inventory of affordable housing stock in Austin become a choice? Let us not make this a battle between urbanists and suburbanists.
Thanks for stopping by and leaving a comment, Larry. Diverse perspectives are welcome here! It sounds like we agree that Austin needs to update its land use policies and you know, pass (!) its affordable housing bond propositions. Certainly, the region’s growth outpacing Austin’s speaks to a housing crunch of some type in Austin. That said, it is certainly a choice to live far away from one’s job. There is no housing commissar in Travis County. As Austinites we can choose to pay to reduce the congestion we experience on our roads by paying for things like a nice suburban commuter rail line and its associated circulator. That seems pretty collaborative (and more regressive) compared to an inbound road toll, for example. The point of this post is that I have no data to help me figure out if the status quo rail line strategy is worth it relative to using that same constrained pool of dollars for other alternatives that might be just at effective on the congestion front and potentially encourage the types of housing density that would support the Austin land use we both seem to want.
Very well done post – a couple of comments:
1. On the “why are we letting Rapid Bus stop light rail on Guadalupe” angle, remember that we probably can’t afford urban rail without the Feds. If the Feds won’t kick in 50% because they think we’re double-dipping on a corridor, then it becomes very relevant. (Not that I agree with this; I’d rather we pay for the first line ourselves, like Houston did, if necessary, rather than going to Mueller because it’s least resistance).
2. On Larry’s post above – this is absolutely relevant, because regionalism cannot just be an excuse for Austin to pay for everybody else’s transportation, leaving our own citizens with nothing but buses. The state already uses our drivers’ gas taxes to subsidize suburbanite commutes; Capital Metro is using our sales taxes at the cost of $22/ride to subsidize rides for suburbanites, most of whom don’t even pay Capital Metro taxes; and now the city’s only option on the table is to use our own tax dollars to circulate Red Line passengers instead of servicing actual living, breathing, Austinites.
One final comment: The “thousands of individuals who provided public input” on the ASG plan in 2004 did so to a plan that had already set Rapid Bus in stone; the public input was about where ELSE they might like to see Rapid Bus in the future. No input was taken about whether rail belonged on Lamar/Guadalupe, for instance. Or whether Rapid Bus was a wise idea given that it might interfere with such.
In 2006, Lee Leffingwell and Brewster McCracken stopped Rapid Bus the first time around for that very same reason: http://m1ek.dahmus.org/?p=350 (excuse formatting; this is one of the entries imported from the old blog I’ve had trouble cleaning up).
Wow, that story on Leffingwell and McCracken is so interesting. Thanks for pointing it out.
You’re welcome – I wish I had saved the Statesman article (live and learn). From memory, basically they said “why are we precluding rail on this corridor for the sake of a service which is basically just running the #101 more often”. Unfortunately, somebody convinced them in the meantime that Rapid Bus was more than the sum of its parts, obviously, because both ended up supporting it the next time around – 2009-10ish.
Excellent work, Julio. One question (at least only one to start): You say the necessary data “is not in the public domain,” but go no further in suggesting the work hasn’t been done at all. What is the likelihood that the studies you reference are all there is?
Hmmm. I would wager that there are a lot more internal analytical documents that have rough comparisons or that explicitly say something like ‘we didn’t even consider the high ridership urban rail because of the politics’. Through all of my reading and following the debate, none of the documents say this explicitly. In public forums, our public transit servants tell a narrative that implies this is indeed the best initial sequence. I am also pretty certain that on someone’s computer somewhere is the modeling software and source data for the CAMPO and 5D-style projections referenced in CATS. So between internal analytical documents and the existing models, we’d have a lot of more insight about what might be optimal. On the other hand, I doubt there’s some super accurate and recent ridership comparison of the proposed circulator against high ridership alternatives based on several growth scenarios for Austin of 2020. For something we might end up spending hundreds of millions of dollars on, I think this is useful information for decision-makers to develop and put out to the public.
Good post. The most recent estimate of daily ridership on the route to Mueller is 9,000-11,000, significantly lower than the earlier estimate for ridership on Lamar-Guadalupe, as you discuss above. An earlier estimate for Mueller was 7,000, and they used several questionable assumptions to bump the estimate up. Whether FTA would accept that is anyone’s guess. I also checked the new recommended breakpoints for the FTA New Starts evaluation, and this ridership level would most likely be ranked “low”. (If there is a high number of transit-dependent riders, they might be able to bump it up to “medium-low”, but I don’t think there’s a lot of transit-dependents on the proposed route, and certainly not compared with a west of campus route.)
Agree with Larry above; not enough affordable housing in town.
Interesting. How did these estimates get computed? Is there a online doc with the methodology and the data used?
Future Connections Study, Appendix B. http://www.capmetro.org/docs/tod/Future-Connections-Appendix-A-C.pdf.
The CAMPO model came up with 5,800 boardings/day in 2017. They add 5,700 boardings, almost doubling this estimate. Their revisions are based on a number of assumptions. I’ll e-mail some details.
Susan, that link is broken.
They must have removed it from the web (hate when they do that.) I have a copy of the report, if you want to provide your e-mail.
it’s chrisbbradford at the gmail thing. thanks
First, let’s not fall into the trap of talking about affordable housing. What matters is household affordability that encompasses housing, transportation, utilities, and basic living expenses. Providing transit options in the urban core will lower the transportation component, and we should not look exclusively at the associated housing costs as the indicator of affordability.
Second, I hope those of you harping on the affordability issue will join me in advocating a dramatic increase in density. We need to saturate all ends of the market (including the high end) so that supply meets demand, and overall affordability improves due to market forces. And please note that holding density hostage by requiring dense projects to include affordable units is counterproductive, as it serves as a “tax” on those projects (while leaving sprawl development untaxed).
I guess affordable housing was not the term I should have used. Workforce housing would be a better term. But I still believe the lack of workforce housing In Austin does not give workers a choice. They are forced out of the market. I will leave it at that. I am a big advocate of density and increased diversity in housing types. Something that will play out as we rewrite our land development code. East Riverside regulating plan,last night, shows a shift in our thinking. A small step but. Step none the less.
I applaud your objectives and approach. The basic engineering approach to define the problem being addressed, evaluate the alternatives and make common sense, logical, cost-effective decisions, based on realistic data and projections, have not been characteristics of the transit program in Austin or in most cities.
This has led to massive wasteful spending which has a disproportionate negative impact on lower income citizens and those who depend on public transit in their daily lives and have no alternative. It makes a mockery of social equity goals and leads to reduced transit service and higher fares for those most dependent. Congestion is not decreased as promised and huge taxpayer transit subsidies for ineffective rail systems provide little (often negative) societal benefits. The current “official” vision of urban rail will increase congestion and substantially increase safety hazards. Its costs are greatly understated and ridership is overstated. The sum of Cap Metro’s sales tax and the City’s property and other tax to support urban rail will make Austin one of the highest transit costs per capita cities in the nation. Likewise, its taxpayer subsidies of transit will be among the highest.
If people are provided true, comprehensive facts with total transparency, they will likely make decisions which will serve the greater good of the community. This was confirmed in the 2000 light rail election.
Austin should not begin with rail as a solution to an undefined problem. It should define the problem to be addressed and honestly evaluate all alternatives. If it is not cost-effective, it cannot be sustained.
Jim, many of us have not forgotten your role in the 2000 election and your misleading ads that almost got you sued by the Statesman, among other things. Kindly buzz off.