Where Cap Metro Jumped the Track

Doug Allen

"I'm here today to talk about the reasons. I'm not here to make excuses."

That's how Capital Metro interim CEO Doug Allen opened his presentation to a committee of the Capital Area Metro­pol­itan Planning Organization at City Hall Monday. The discussion was listed on the Transit Working Group's agenda as "MetroRail, lessons learned."

The major lesson? Don't put the cart before the horse.

"The budget and the timeline that was established was set way too early and before the project was fully defined," Allen told the TWG, a rail advisory committee to CAMPO, the group of elected officials that guides transportation planning in the Austin area. "That's no one's fault but Capital Metro's."

The result, said Allen, is that Capital Metro presented the public in 2003 with dollar figures and an estimated starting date for the Leander-to-Downtown commuter rail project – to convince voters to approve it in Novem­ber 2004 – but after the election, the transit agency quickly discovered that it didn't know what it was doing. Those missteps ultimately led to the technological snafus that have delayed the rail starting date from the originally predicted 2008. The debut is now projected for no sooner than the first quarter of 2010, and it will cost $105 million, not the $94 million initially stated (or the $60 million originally on the ballot, which did not include the cost of leasing rail cars).

"I think early on, the complexity of the signaling system may not have been appreciated," said Allen. "The staff needed to manage that effort was probably not where it needed to be. ... Rail system expertise was not fully prevalent at the beginning of the process. There was some rail expertise brought on to the project in 2006. Primarily, that expertise was from a light rail environment and not a commuter rail environment. Some of those technologies are different." (Allen, it's worth noting, wasn't hired as an executive vice president until March 2008, after 14 years at Dallas Area Rapid Transit.)

Instead, Allen said, the agency should have followed what he said was a more typical course: First develop a regionwide system plan, perform a corridor study, define the project, and then design the project (and during design, start formulating a budget). After that, begin construction, and then perform final testing and validation. The total duration, he said, should be from 7.5 to 10.5 years, significantly longer than the four years attempted with MetroRail.

"The system actually went to construction before the design was completed, in order to maintain that timeline," Allen said. Ironically, he noted, that mandate to stick to the timeline was to establish some credibility for the agency after years of mismanagement back in the 1990s.

Austin state Sen. Kirk Watson – chair of CAMPO and ex officio member of the TWG – said a project of this type should first establish the "who" (who has the expertise) before defining the "what" (project definition) and "when" (the timeline). "After the referendum, you brought in people with experience. That violates the who, what, and when," he said.