Are HOV lanes a dead end? There's evidence they're not relieving congestion

Tuesday, April 17, 2001

By LARRY LANGE SEATTLE POST-INTELLIGENCER REPORTER

So you think you can zoom along in your carpool lane, rarely dipping below a comfortable 45-mph speed?

It's supposed to work that way. But increasingly, the state's high-occupancy-vehicle lanes are caught in the same congestion they're designed to prevent.

On several recent afternoons, for example, our spins down Interstate 5's diamond lane screeched to a halt around Southcenter. Northbound carpooling, too, fell short of the 45-mph standard most of the time from downtown Seattle to the Snohomish County line.

For more transit stories, Webcam links and previous Getting There questions, visit the P-I's Transportation Page. In an era when building new freeways has been politically and financially off limits, the state has relied heavily on HOV lanes to relieve the Puget Sound area's growing traffic headaches.

Yet, in the three decades since Washington began developing HOVs, it has done little tracking to determine how well the $1 billion system works.

And there are new signs that the lanes are beginning to lose some of their effectiveness.

A Seattle Post-Intelligencer analysis of the limited state information available found several surprises:

To meet state standards, HOV lanes should move at 45 mph or faster at least 90 percent of the time. But in many cases that isn't happening.

"It's a topic we're just starting to talk about," said state traffic engineer Matt Beaulieu.

The state has consistently monitored only two locations along the 191-mile HOV network, which connects crowded areas from South Snohomish to South King County. Even in those two spots, at Hunts Point and Shoreline, tracking of HOV effectiveness dates only to 1994.

The whole point of HOV lanes is to squeeze more people into fewer cars. But the trend in both of the monitored spots is going the other way -- the average number of people riding in each vehicle has declined since 1994. The total number of people moved in the lanes, meanwhile, has fluctuated.

Building HOV lanes hasn't stopped traffic from growing on adjacent general-traffic lanes, where the worst congestion is. Four years of I-5 traffic counts show that, from SeaTac to Everett, general traffic often kept on growing from year to year even as HOV use increased. Just 31 of 315 traffic readings showed HOV use up coinciding with other traffic down.

This year will be the first time the state will analyze trends in the lanes' performance.

State Sen. Dino Rossi, R-Issaquah, is no fan of HOV lanes. He wants them opened to all traffic between rush hours.

But the HOV system remains popular in Olympia. House Transportation Committee co-Chairwoman Ruth Fisher, D-Tacoma, thinks the only issue is finding the money to complete the system. She said it relieves traffic even on weekends.

The system does relieve traffic by putting more people into fewer cars. For example, the HOV lane on westbound state Route 520 at Hunts Point carries almost as many people during the peak morning hour as each adjacent general lane, but in almost 90 percent fewer cars.

On Interstate 5 at Northeast 145th Street, vehicles in the HOV lanes carry nearly three times as many passengers as those in other lanes. And the number of cars, buses and people using the lanes has grown.

The lanes are "doing exactly what they're supposed to do," said Mark Hallenbeck, director of the Transportation Research Center at the University of Washington.

Slower speeds

But speeds in some lanes have slowed as use increases and motorists worry about safety. Hallenbeck, who studies their use for the state Department of Transportation, said traffic on the system frequently dips below the desired 45-mph speed.

"I remember being surprised that two of us, even with access to the car-pool lane, weren't able to buzz through" on I-5 at Southcenter Hill, said Beaulieu, the state engineer. The movement is "still very significant. ... But it's not free flow," he said.

By 1999, HOV speeds had dropped below 45 mph as often as 30 percent of the time on southbound I-5 between Northgate and Alderwood. They also fell short 25 percent of the time on Interstate 405 north of Interstate 90, and 15 percent of the time on I-5 south of downtown Seattle, according to one study.

Officials attribute this to motorists driving at slower speeds to avoid hitting other vehicles pulling into the lanes.

We know about HOV speeds. We tested them. The P-I's own HOV jaunts found that, overall, the lanes were faster. Indeed, we zipped past I-5 traffic jams until ...alas, carpools were jammed too.

Our goal: to find out if I-5 HOV lanes met the state standard of going 45 miles an hour 90 percent of the time. On three afternoon drives from downtown Seattle to South 188th Street, we hit the target as little as 23 percent of the time and never more than 66 percent of the time.

Northbound between Lake City Way and Mill Creek, we traveled at less than 45 mph for 75 percent of the time, although we passed other cars.

Drops in HOV use

Meanwhile, there are signs that HOV use is dropping in some places. On 520 at Hunts Point, the westbound lanes averaged more than 13 people per vehicle during the peak morning hour in 1994. In 1999, the average number of people per vehicle was down to 11.6, though it rose from 1998.

That average is vastly greater than in adjacent lanes, where vehicles usually carry only the driver.

On I-5 at 145th, the per-vehicle average in the HOV lanes dropped slightly between 1994 and 1999.

Hallenbeck attributes the declining vehicle occupancy to growing numbers of HOV lane users in carpools, which carry fewer passengers in each vehicle than buses.

The growing Eastside job market means "there's a slight decrease in the number of people trying to go (west) in the morning," he said.

By one measure, HOV performance is improving: Since 1998, the total number of people moving in HOV lanes at peak hours has increased at most of the five points measured.

Carpools up, loners up too

The state's most detailed data on HOV use is in the Department of Transportation's ramp and roadway traffic reports. But they, too, offer an incomplete picture. Hallenbeck said they are taken directly from traffic-counting devices that occasionally stop working or produce questionable readings.

The P-I examined four years of the traffic counts to see if traffic shifted from crowded general lanes to more freely flowing HOV lanes.

HOV use did grow. But only in 31 of 315 counts did general traffic decline at the same time. And the state is not sure whether drivers actually shifted to carpool lanes or simply stopped driving in the crowded lanes. Congestion may slow traffic enough that fewer cars get counted.

Because officials don't query drivers about why they shift from one lane to another, they can't readily explain.

Traffic changes also reflect some drivers' change in commuting routes.

"I don't think you can say it's any one thing," said Leslie Forbis, a state HOV operations engineer.

Gaps in information

Searching for trends in the HOV numbers is complicated by lack of information. For example, two recent reports profile five HOV locations, but a similar 1995 report looked at mostly different locations.

State officials say locations changed because of construction activity on Interstate 405. Also, budget limits restricted the number of measuring points.

So how can the department say that "HOV lanes are good for both the highway and transit systems?" Officials said it's because even the limited data they keep consistently shows HOV lanes carry more vehicles and passengers in a peak commuting hour than other lanes.

"It makes a huge difference in the commute from Everett to Seattle," said state Transportation Commission chairwoman Connie Niva. Speaking on a cell phone from her car while stuck in a recent traffic jam, she said: "The HOV lane (traffic is) flying and I'm just sitting here".

A state study, due for release this summer, may answer some questions. It will analyze numbers of vehicles and people using HOV lanes as well as speeds traveled. It will compare those figures to similar ones gathered two years ago at the same locations and attempt to reach some conclusions.

-- Mark Stilson (mark842@hotmail.com), April 19, 2001