Someone recently asked me what I thought were the nation’s worst-managed transit projects. I suggested the Honolulu rail was number 1, the Maryland Purple Line was number 2, and BART to San Jose was number 3. But maybe I underestimated the insanity of the BART-to-San Jose line.
Estimates for cost and completion date of the BART rail extension into downtown San Jose:
2014: $4.4 bln, 2026
2020: $6.9 bln, 2030
2022: $9.3 bln, 2033
2023: $12.2 bln, 2036
2024: $12.8 bln, 2037
?The project has yet to break ground.— John Arnold (@JohnArnoldFndtn) March 12, 2024
It’s even worse than Mr. Arnold suggests. In 2001, the Santa Clara Valley Transportation Authority (VTA) did its initial alternatives analysis comparing BART with a wide range of alternatives including buses, bus rapid transit, commuter rail, light rail, and “Diesel light rail,” which is what the FTA now calls “hybrid rail.” BART was picked because they thought it would get the most riders even though it was also by far the most expensive.
Although it was well over twice the cost of the next-most-expensive alternative, the projected cost of building BART from Fremont to downtown San Jose was just $3.7 billion in 2001. That included the portion from Fremont to Berryessa, which was nearly two-thirds of the miles but (because it was all above ground) was expected to be just a third of the costs. That portion has been completed and the costs quoted by Arnold above are just for the portion from Berryessa to downtown.
That means the 2001 projected cost from Berryessa to downtown was only about $2.5 billion. In today’s dollars, using a standard inflation index, that’s about $4.4 billion; using a construction-cost index brings it up to $5.6 billion, still well under half the current projection.
In addition to underestimating costs, it is highly likely that the 2001 analysis greatly overestimated ridership. We won’t know for sure until and unless the project is ever completed, but in 2001 the planners projected an average of 87,200 riders per weekday. By 2011, this had dropped to 57,400, and the post-pandemic projection is for just 32,900 weekday riders.
“The systematic tendency to over-estimate ridership and to under-estimate capital and operating costs introduces a distinct bias toward the selection of capital-intensive transit improvements such as rail lines,” US DOT analyst Don Pickrell pointed out in his 1990 review of predicted vs. actual rail transit construction costs. “This bias arises because, as a variety of studies has shown, rail becomes the economically preferred transit-mode only when its substantial capital costs and fixed operating expenses can be spread over large passenger volumes.”
BART didn’t make sense even in 2001. Both the capital and operating costs of the BRT alternative were less than a third as much as BART yet that alternative was projected to attract more than half as many riders. If, instead of blowing $3.7 billion in one corridor, VTA had introduced BRT lines in three major corridors, it could have gained 70 percent more riders at a lower operating cost than BART. Such BRT lines could have been running 15 or 20 years ago with relatively negligible cost overruns, while BART is taking more than 35 years from alternatives analysis to opening.
In 2005, the Bush administration imposed rigorous cost-effectiveness criteria on transit projects for the first time. Congress, which had included cost-effectiveness language in the law since 1991, immediately responded by exempting BART and three other projects — the DC Silver Line, Portland’s Westside Express, and San Francisco’s Central light-rail subway — from the Bush rule. All of those projects have turned out to be horrible disasters: construction of the Silver Line forced Metro to cut service on the Blue Line, which cost the agency more riders than the Silver Line gained; the Westside Express went 60 percent over budget and never attracted more than two-thirds of its expected ridership, at one point costing taxpayers well over $100 per passenger; the San Francisco light-rail subway was originally supposed to cost less than $450 million but ended up opening four years late at well over three times that cost.
VTA can save taxpayers billions by shutting down its BART project, but it won’t. It knows that Congress is in a spending mood and that no one who is spending that money cares whether the new line carries any riders at all.
