Why is everything late, over-budget, and broken in San Francisco?

San Francisco is building big things these days, but not always particularly well.

The Transbay Transit Center, which finally reopened for actual transit traffic over the weekend, cost far more than it was supposed to and finished late. It then almost immediately broke, and spent most of a year shuttered.

Similarly, the Central Subway, still under construction, is already more than a year late and may not actually carry passengers until May of 2020.

Muni’s much-anticipated new trains appeared late and, when they finally did go into action, proved a hazard to some riders. BART’s own “fleet of the future” lingers far behind schedule as well. And the less said about the state of the state’s high-speed rail project, the better.

Whether it’s our sinking towers or our rusting bridges, it seems that nearly everything built in the Bay Area these days—and particularly big civic projects—is broken one way or another. Why does this keep happening?

Over the years there’s been quite a bit written about the perpetual problems with “megaprojects,” those big-ticket ideas that cost $1 billion-plus, have the potential to reshape cities, and almost never work out the way they’re supposed to.

This is all theoretical work, but certain commonalities tend to emerge, including:

• It’s not just SF; megaprojects are almost always disasters: In 2002, Danish economist Bent Flyvbjerg and associates published an infamous paper surveying 258 megaprojects around the world that found that some 90 percent of them came in late and over budget. “The cost estimates used to decide whether such projects should be built are highly and systematically misleading,” Flyvbjerg concluded, warning municipalities in blunt terms “[do] not trust cost estimates and cost-benefit analyses produced by project promoters and their analysts.” He felt so strongly about this that in later writings Flyvbjerg coined what he called “the iron law of megaprojects: over budget, over time, over and over again.”
• Low bids are not necessarily good bids: By law, the state of California must award construction contracts to the lowest bidder in almost all cases, and many cities employ similar policies. This is supposed to save money, but what it actually does is incentivize contractors to make unrealistic promises to nab the deal. Companies like Tutor Perini are expert at playing the system to get hired, but less so at actually delivering—just that one company has gone over budget by $765 million on Bay Area projects since 2000.
• One problem feeds into another: New York-based consulting company McKinsey & Company’s 2015 article on megaprojects observes that once unrealistic expectations are in place, the pressure to (somehow) deliver creates future hazards. “The temptation is to cut corners to maintain cost assumptions,” which can lead to “incomplete design, lack of clear scope, ill-advised shortcuts, and even mathematical errors in scheduling and risk assessment.” And this is exacerbated by the scale of the projects, because “what might seem like routine issues” on any other job become exponentially worse with these.
• Extreme problems become more common at larger scales: Similarly, problems can become deceptively more likely as goals get more ambitious. Writing for the Cato Institute (a libertarian think tank previously known as the Charles Koch Foundation) in 2017, Flyvbjerg says that “delivery is a high-risk, unpredictable activity, with overexposure to so-called ‘black swans’—that is, extreme events with massively negative outcomes.” But “managers tend to ignore this,” resulting in what we might think of as the Jurassic Park model of people treating extraordinary risks the same way they would lesser ones.

Photos by Robert Pierce

• “Weak leadership”: Also while writing for Cato, Flyvbjerg points out that “often projects are led by planners and managers without deep domain experience who keep changing throughout the long project cycles that apply to megaprojects, leaving leadership weak.” For example, the Transbay Joint Powers Association formed in 2001 with a five-member board; 18 years later the body has changed personnel many times, and today’s eight-person board retains no original members, nor would anyone really expect it to. This doesn’t necessarily mean that members are weak individually, but it doesn’t help that those who aren’t won’t necessarily be there for long.
• Politics makes everything worse: Given the nature of these proposals and the amount of money involved, it’s impossible to keep politics out of them, which almost never helps. University of Alberta Professor Allan Warrack’s 2013 paper “Megaproject Decision Making” poses that political systems don’t incentivize honest evaluations of big construction ideas, arguing that “negative decisions are easier to make (and justify politically) than positive ones” and that “easier still are delays in decision-making,” which make bad decisions later on more likely thanks to the pressures of putting it off. In other words, politics may simply make it too tempting to tell people what they want to hear.
• It’s possible cities need to be wrong to make progress: It sounds bizarre, but in his influential 1967 book Development Projects Observed, economist Albert O. Hirschman suggested that if people realized the true cost and risks of megaprojects they’d probably never agree to them—and nothing of significance would ever get done. Hirschman called this his “hiding hand principle,” writing that society “takes up the problems it thinks it can solve, finds out they are really more difficult than expected, but then, being stuck with them, attacks willy-nilly,” a process that is ultimately constructive in the long run.

Hirschman’s notion is not necessarily a popular one, and some critics resist it for fairly obvious reasons.

But it does explain why the same problems seem to keep perpetuating themselves over and over again—if the only realistic alternative is to remain static, there’s almost nothing else that can happen.