Neil: BART to downtown San Jose will cost $12.8B. How did we get here?

VTA’s BART to Silicon Valley Phase II will build four new BART stations over six miles, extending from Berryessa/North San Jose to Santa Clara.

The project has been subject to a lot of criticism. It’s been repeatedly bashed for a redundant Santa Clara station, choice of a single-bore design for the tunnel and its $12.8 billion price tag. The discussion around these issues has lacked necessary context and amplified the voices of opponents without a word from the project’s supporters.

BART SVII isn’t flawless, but it’s important to know the limitations imposed on it that have caused the issues many are concerned about. As a civil engineering student, transit advocate and nerd, I’ll try to set the record straight.

Santa Clara station has been a target for many who speak about the project, citing the redundant connection to Caltrain and the need to cut costs. While logical, these calls ignore the fact that VTA will also be building the Newhall Yard & Maintenance Facility, a necessary addition for the extension.

A few hundred feet of track and a platform are a small price to pay for an estimated 10,000 riders per day by 2040. And since the trains will be running to the Santa Clara yard anyway, why not build a station?

VTA considered two construction methods for the project: a tunnel boring machine, which digs underground like a mechanical earthworm, or cut-and-cover, where you dig down and cover it over. Neither is perfect and both have trade offs. Tunnel boring machines reduce surface disruption at higher costs. Cut-and-cover increases disruption for lower costs.

San Jose and VTA, drawing on the experience of disruption from previous surface projects, ruled out cut-and-cover. Right or wrong, that was their reasoning. But why one bore instead of two? According to VTA, a twin-bore design uses cut-and-cover for stations, which could also cause conflicts with utilities that would be avoided by a single-bore.

The decision to choose a single-bore was short sighted, but I can understand VTA’s reasoning. I would prefer a single-bore project built as soon as possible over a twin-bore project that would require unknown delay to redesign everything, resulting in cost increases.

This project is expensive because of  inflation, high wages, mismanagement, single-bore design, labor shortages and consultant costs. If it were twin-bore, would it be cheaper? Maybe. Cut-and-cover? Maybe. Managed better? Maybe.

The elephant in the room is that America can’t build inexpensive projects. U.S. infrastructure was never cheap, but since 2020 costs have skyrocketed. The Federal Highway Administration keeps an index of construction costs which paints a grim picture. While BART is not a highway, the index is still useful. Since Q2 2020 when the extension was estimated at $6.9 billion, costs increased by 60% to Q3 2023. Using that number, the project should cost $11.05 billion. We’re 15% over that, which is cause for concern, but not as disastrous as it seems without context.

BART SVII is imperfect, and with hindsight there could’ve been changes, but it’s not the only project seeing cost increases. We’re a victim of a national trend. VTA should be working to cut costs and build the project quickly, but it’s not VTA’s job to fix a national problem. High costs for big projects are always painful. Let’s focus on getting it done.

Harry Neil is a civil engineering student and transit advocate from San Jose who works with SCC4Transit. He is not affiliated with any government agency.