On November 8, voters rejected elevated rapid transit in Seattle. By the next day, the city had begun a debate about what sort of transit system would rise in its place. The idea that arose most prominently, in a November 13 op-ed in the Seattle Times, was a "solution" that has long been a dream of road supporters: A 150-mile network of so-called "bus rapid transit" lanes that would "cover the entire metro Puget Sound region at 60 mph, 24/7"; "support 'walkable' mixed-use neighborhoods"; carry far more riders than light rail or monorail; and include "clean... comfortable stops" that are "the equivalent of rail stations"—all at a tenth of the cost of a fixed-guideway system like the monorail.

Promises like these have been alluring cities since at least the 1960s, when bus manufacturer General Motors began aggressively pushing BRT as an alternative to rail. And, despite lingering concerns that buses don't offer the same stability and psychological appeal as trains, BRT's popularity has endured. Last week, at a post-monorail transportation forum sponsored by the Sierra Club, Seattle City Council Member Richard Conlin argued that BRT might be a viable alternative to monorail in the Ballard-to-downtown-to-West Seattle corridor. "My gut reaction is that it might work really well," particularly in West Seattle, Conlin said this week.

But the realities of BRT have rarely lived up to its promises. In city after city—despite assurances that BRT would offer cheap, flexible, speedy transit service—the technology has proven costly, inflexible, and anything but rapid.

One problem in addressing the drawbacks of BRT is that no one, including its backers, seems able to agree on a single definition of the term. Some apply it only to grade-separated roadways (like the downtown Seattle bus tunnel, or Sound Transit's elevated "E3 Busway") that are reserved exclusively for buses. Others use it to describe a broad range of services that includes enhanced express bus service on HOV lanes, buses that move from dedicated bus lanes to mixed traffic, and express buses between the suburbs and the inner city. "If you want real BRT, you have to be talking about exclusive right-of-way, exclusive rail-type stations, and real land-use changes around those stations to accommodate growth," argues Rob Johnson, policy director for the pro-transit Transportation Choices Coalition.

Many transit agencies have adopted a much more flexible definition. One of the largest and most commonly touted U.S. examples of "BRT" is the Silver Line in Boston, essentially a bus line that runs on surface streets that have been converted for bus-only use.

The primary argument for BRT, especially during the Bush era of parsimonious transit funding, is that it's cheaper and easier to implement than light rail. But while it's undeniably less expensive to put buses on existing streets than it is to build the substantial infrastructure needed to create a new rail transit system, there are other measures of cost-effectiveness besides capital costs.

For example: How many transit riders will a new BRT line draw? How many of those were already riding buses anyway? How much can a city expand the capacity of BRT before the buses start slowing down? How much more investment capital would a fixed-rail line have brought to an area? And what kind of subsidy has the government already given to build the roads on which BRT buses run?

On the first two questions, the data is clear: BRT draws far fewer transit riders—and, importantly, far fewer new transit riders—than light rail or other fixed-rail systems. In a 2001 study that's often cited as evidence that BRT can work along the former monorail Green Line, the Seattle Department of Transportation found that elevated transit like the monorail or elevated light rail would add about 56,000 daily riders to the North Seattle-to-downtown corridor; BRT would add just 32,500. From West Seattle to downtown, the disparity was even more startling: nearly 28,000 riders for elevated rail, and just 10,000 for BRT.

Real-world statistics bear out the Seattle planners' estimates: In Houston, which recently dropped a voter-approved plan to add 13 miles of rail to its Metrolink light-rail system in favor of new "rapid" bus service and commuter rail to the suburbs, there are six BRT routes running on 44 miles of freeway HOV lanes throughout the city. Currently, just 36,000 people use the system. In Portland, a much smaller city both geographically and in terms of population, a 33-mile light-rail system carries nearly twice as many riders as Houston's: some 74,000 a day. Because of the higher ridership, the cost per passenger mile—a common measure of cost-effectiveness—is actually lower in many cities, including Portland, for rail than it is for "affordable" BRT.

Another downside: Bus lanes, unlike rail, can be easily converted for use by other types of vehicles, in effect subsidizing private autos with public-transportation dollars. In Houston, highway lanes that were originally dedicated to "bus rapid transit" have been converted into HOV lanes where buses compete with private cars. This is exactly why you'll never see real economic development around a bus stop: Buses can be moved; trains have to go where the rails go. But the most obvious argument for rail over BRT may be that it is empirically, undeniably, faster than riding the bus. According to the city's 2001 transit study, riding an elevated train from Lake City through Ballard to downtown would take half an hour less than taking the same trip on a "rapid" bus. From West Seattle to downtown, an elevated trip would take 20 minutes less than it would on BRT: a total time difference of nearly an hour from one end of the city to the other.

The reasons for the glaring disparity will be obvious to anyone who rides the bus. Buses, like at-grade light rail or streetcars, get stuck in traffic. Even sometime BRT proponent Conlin acknowledges that in order to build truly rapid busways, you have to elevate them or put them underground. At that point, he says, "there isn't much point in doing it for a lower-capacity mode like buses, when you could do it for a higher-capacity mode like rail."

Like, say, the monorail?