"It'll help reduce congestion on one of the busiest freeway corridors in the region," Mayor Antonio Villaraigosa said at a recent press conference warning drivers to stay away during the closure. "There's an obvious long-term gain, but there will be short-term pain."
The benefits, however, are far from obvious. In fact, there is growing consensus among those who study transportation that an extra carpool lane will not appreciably reduce congestion over the long term. Instead, it will simply put more cars on the road.
"You will see in the short run there is a release in congestion," says Allison Yoh, associate director of the Institute of Transportation Studies at UCLA. But over time, she says, drivers will adjust their schedules and routes and traffic will fill in again. "In the end you are back to the same level of congestion you were at before," she says.
Of course, she would say that. She's a public transit supporter. But that conclusion is shared even by highway supporters.
"The jury is in," says James Moore II, director of USC's transportation engineering program and a highway proponent. "Nobody who is in the business of designing transportation systems or building transportation systems believes that adding capacity is going to alleviate congestion in the long run. That capacity will be used."
Well, not quite nobody. The Metropolitan Transportation Authority and Caltrans, who are spending $1 billion on the Sepulveda Pass project, vigorously defend the project and maintain that it will indeed alleviate congestion. But their case consists largely of outdated talking points, based on questionable metrics and dubious assumptions.
At his press conference, Villaraigosa noted that the project will close the final gap in the region's carpool network, providing a continuous carpool lane on the 405 from Orange County to the San Fernando Valley. Commuters will save "one minute per mile in the carpool lane," he said. "That's 10 minutes per day saved in the Sepulveda Pass."
The minute-per-mile figure has been frequently cited over the years. But as carpool lanes have become more and more congested, Caltrans has backed away from it, says Marco Ruano, Caltrans' director of freeway operations.
"I don't know that we're using that as much anymore," he says. "It's almost like the HOV lanes — in some areas of the region, they're a victim of their success."
So what tangible benefit can carpoolers expect?
"Overall we feel there will be benefits for users, but I can't give you a number," Ruano says. "I can't say you'll save 20 minutes if you leave the Valley and come back. There will be improvement."
There are other ways to answer the question, but they aren't any more reassuring. One measurement is "level of service," which is a letter grade based on average traffic speeds. A Caltrans study showed that during rush hour in 2003, segments of the northbound 405 rated mostly E's and F's.
So would adding a carpool lane improve those failing grades? No. The study showed that the Sepulveda Pass project would boost traffic volume while reducing the level of service to straight F's by 2031, mostly due to anticipated growth.
"There will be an immediate benefit," says Doug Failing, the executive director of MTA's highway program. "But when you look into the future, it'll probably crawl back up to level of service F again."
Ruano argues that without the project, the freeway would fall "several levels down from F" — if such a grade existed.
Failing argues that a better way to see the project's benefits is to look at overall "vehicular delay." A 2006 traffic study concluded that the project would reduce total delays by 15,000 hours per day by 2015. But the study makes the explicit and wrongheaded assumption that adding freeway capacity will not increase demand. Assuming otherwise "would require regional modeling that is beyond the scope of this study."
If the model can't account for reality, the results aren't worth very much.
"We see that all the time," says Damien Newton, editor of Streetsblog L.A., a transportation advocacy website. "They all assume that adding capacity is not going to add demand."
On his blog, Newton has been skeptical about the wisdom of the project. He argues it may create more gridlock during construction than it alleviates, and has suggested that money might be better spent on public transit. He also has noted that carpooling has not increased even as L.A. County has built one of the nation's most extensive carpool systems.
Caltrans hotly disputes that, but Newton is backed up by the Southern California Association of Governments, which has concluded that the vast investment in carpool lanes has failed to change travelers' behavior. The agency's most recent transportation plan states that building out the system has not resulted in more carpooling as a share of overall traffic. Martin Wachs, a principal transportation researcher at the Rand Corporation, noted that the vast majority of "carpools" are actually members of the same household.
Carpool lanes were all the rage in the transportation world about 25 years ago. Given the slow pace of bureaucracies, they remain a high priority for transportation agencies. Among researchers, however, they have fallen out of favor.
Now the hot topic is toll lanes. With congestion tolls, drivers could choose to sit in traffic or pay $2 or $4 or $10 for a congestion-free ride in the carpool lane.
"If you ask most people who study congestion professionally what we should do, they usually say you need tolls," says Michael Manville, a researcher at UCLA. "For various political reasons, that's not something policy has caught up with."
Charging drivers to use a freeway is a tough sell. But where there is high demand, freeways cannot be both free and free-flowing, researchers argue.
"The roads are scarce. ... If you don't charge for it you're going to get queueing," Manville says. "You see it every morning. For whatever reason politically, we have decided to have Soviet-style freeways."
Despite stiff resistance to the idea, the MTA has taken some tentative steps in the direction of congestion pricing. The agency plans to launch a pilot project on the 110 freeway in the fall of 2012.
Moore, a staunch supporter of congestion pricing, suggests that completing the carpool network through the Sepulveda Pass will help reduce congestion if MTA is planning someday to convert the carpool system to tolls.
But even if it doesn't, Moore argues, that doesn't make the project a waste of money. It will add capacity. That will mean more economic activity — more trips to school, to work, to the dentist. All of that is good. Commuters just shouldn't expect to get there any faster.
"If you're spending money to add capacity to a very congested facility, the benefits have always swamped the costs," Moore says. "The reason people are surprised that there aren't congestion improvements is that we keep telling them there will be. And we should stop."