Artwork by Oran, of course
Artwork by Oran, of course

Last month, in one of several posts on Southeast Seattle service changes, I mentioned Metro’s attitude towards frequent rider requests for a local circulator:

Furthermore, after to talking to some of the Metro planning staff, it’s an article of faith in some quarters there that “circulators don’t work.”

My throwaway remark prompted a comment from Jarrett, linking to his extended response on his blog, “Human Transit”.  He critiqued my circulator concept as a way of debunking the entire idea of circulators. In his comment thread a bunch of Metro commenters form an amen corner, pretty much proving the limited point I made above.

Jarrett’s piece, all the way from Australia, is eloquent and informative. Some key grafs:

Forgive me if this seems obvious but: People get on a bus because it takes them to (or at least toward) where they’re going. The shorter a route is, the fewer places it goes, and thus the fewer people will tend to get on it. Very short routes, say under 3 mi, tend to do very poorly unless they’re the sole means of reaching a major destination (such a shuttle between a major hospital and a nearby rail station.) Metro’s own experience with a circulator route in West Seattle is a good example.

If a neighborhood is served only with a circulator, this also puts the area one connection further away from most destinations. Metro’s proposed structure keeps the Rainier Valley connected to the rest of the city by continuous north-south lines, increasing the likelihood that only one connection will be need to get most places…

There’s a much broader lesson here that I’m sure I’ll come back to. Localized interests (such as developers, community groups, and small city governments) tend to invent short routes — “circulators” or “shuttles” — because they are thinking about their particular transport problem in isolation. Successful transit agencies, though, are always looking for how to serve these localized needs using longer routes that do many other things. It’s by combining markets, not by serving them separately, that successful transit corridors are made.

I think I’d concede the larger point Jarrett is trying to make. The classic circulator, which takes a scenic route through the neighborhood before getting to the transit station (or what have you), is a terrible route. First of all, it doesn’t really take you anywhere you want to go beside the station; secondly, due to its milk-run nature, it’s actually a pretty poor way to get to the station if you have any alternative.

In spite of all that, I think the “double loop” concept I put on the table (see map above) is not that classic circulator, because of transit usage in the Rainier Valley, the non-trivial geographic span of the proposed lines, and certain dynamics of travel within the valley:

  • As has been pointed out many times, in the Rainier Valley buses are as much a way to complete mid-day grocery shopping as get to work. If a route serves the main retail centers, it will attract riders irrespective of its connection to light rail.
  • The double-loop configuration allow rapid travel along the key Rainier Ave retail corridor and the heavily populated Beacon Ave corridor, while also ensuring reliable, quick connections to the nearest rail station, which in time will become important business and residential concentrations in their own right.
  • Most importantly, the circulators enable seemingly obvious trips that are totally impractical today. Someone living at a random point on Beacon Ave attempting to reach a random point on Rainier Avenue, today, is typically looking at a three-seat ride. For obvious reasons, the two corridors have become silos where bus-mobile residents rarely interact, in spite of the fact that they’re perhaps a mile from each other.

A common theme during the service change process was “East-West connections.” I think Metro understood that as new ways to get out of the Southeast without going downtown. While that’s important, they neglected the fact that it is very hard to go East-West within the valley, unless you happen to live on one of two or three widely dispersed cross streets.  The current service plan doesn’t do much to change that, although it does preserve parallel service to downtown 3 or 4 blocks from a light rail line.

30 Replies to “More On Circulators”

  1. Is there stop-level data on usage? That seems like a good place to start. Perhaps this could even be a pilot study using ORCA data if there’s enough available yet.

    Also, according to a “graf” is a German noble. What does graf mean to you? :)

    1. Graf is a count. In French the word is comte, and the land he reigns over is a comté. The Normans applied this word to English districts (shires), which became county in English.

      I wrote in the zoning article yesterday about circulators, and how the 39 is an important one that needs more frequency. People will ride a circulator if it’s coming in five minutes. They won’t ride it if they have to fit its 30-minute or 60-minute schedule. That’s why the West Seattle circulator failed, as well as the downtown Bellevue circulator in the 80s (Bel-hop).

      They are right that a bus should ideally “go” somewhere, and not just to a transfer station. A corollary to this is a bus should have a destination at both ends. The 30 does, because both Magnuson Park and Seattle Center are attractions. The 15 doesn’t; it just peters out in a residential area, so all the ridership is one way. ‘Course that’s not the bus’s fault, since there is no destination that direction. But the point remains that the routes with the most ridership have a regional destination and both ends and in the middle.

      This double-loop proposal is an interesting idea. A perfect route for vans.

      1. I disagree that the bus must have a destination on both ends. Most bus routes function fine as routes to get people from their homes in residential areas to destinations.

      2. I agree with alexjonlin that there are good and useful routes that don’t end in destinations.
        For example, take route 14 (the closest to my house). On one end, it goes to west Capital Hill. It then visits downtown, the international district, and the central district. However, its south terminus is no more than a small neighborhood. Even though it doesn’t have major destinations at either end, it is usually standing room only for much of the route and routinely gets more trips added at service revisions.

      3. Everyone gets so defensive whenever anyone criticizes a bus route. Route 14 would be a better route if instead of turning around at Belmont it instead continued north to the university. While the 14 has great ridership, it is practically always empty at the “end of the line” up on Capitol Hill. The 14 is also packed going into downtown in the morning and empty coming back, then packed leaving downtown in the evening and empty going back. That’s a waste of resources. I can’t speak to the southern end, but if the 14 continued north to the U-district via eastlake, then it would be carrying passengers in both directions at all times of the day, doubling the effectiveness of the route.

        To say that a route ideally should begin and end at a “destination” does not mean that any route that does not begin and end at a destination is a bad route; it just means that a good route could be a better route if a reasonable way can be found to make it bi-directional.

      4. I agree completely that the north end of the 14 is a waste of resources. I think it would be better if it terminated downtown, both for reliability’s sake and because very few people who ride it in from the south continue on as it turns east. Also, the 43 provides service to exactly the same places (and more!) as the 14 east of downtown.

        The southern half of the route is consistently high-ridership. buses are usually standing-room only all the way up to 23rd & Jackson, and there are always plenty of people who get off for the rest of the route. In peak times especially, many full 40-foot buses run from Jackson to the southern terminus.

      5. The 14 is a leftover from the streetcar days. Who knows what they were thinking when that route was planned.

      6. I think the south half of 14 is good, but the east/northern half is a completely unnecessary duplication of the 43.

      7. The Summit loop portion of the 14 does have a fair number of riders and passes through a neighborhood that is almost entirely multifamily buildings. The area likely has such a large number of multifamily properties due to being along a streetcar line.

      8. The #14 Southbound route takes you from dwtn to 23rd & Jackson to schools to Leschi to M Baker – and in Sept will connect to the Mt Baker Light Rail Station. This is a critical destination!!

      9. Also worth noting on the #14, is that the Summit Loop is a reason many people live in that neighborhood, it’s low ridership mid-day, when the buses come every thirty. But, living in that neighborhood, you see every stop on the Summit Loop with a good size group of people in the morning. I’d vote for increasing the mid-day service enough the transit-friendly neighbors of it, would ride it even more. (You will see many people check the schedule, then walk downtown if nothing is coming.)

  2. 1. Circulators work if you’re trying to provide coverage in an area with minimal cost. It’s something you do when there isn’t enough money for 2-way routes.

    2. I love when people talk from a cartography standpoint, without observing the physical streets they are referring to.

    This is predominately moot because as I said before, Metro is BROKE!

    Brian Bradford
    Olympia, WA

    1. I concur that uni-directional loops are a terrible idea and I’m not sure if that is what martin is proposing here or not, but a bi-directional loop, where buses run in a circle, but in both directions, can be an excellent configuration if the land use pattern demands it.

      Imagine a BI-DIRECTIONAL loop that goes in a circle from residential zone to neighborhood commercial zone to a residential zone to another neighborhood commercial zone, to another residential zone to yet another neighborhood commercial zone and then back to the place it started. Anyone trying to get from any one of those residential zones to any one of those commercial zones would be served via a single-seat ride along the most direct route possible. The route would be free of deadheading connecting origins and destinations all along the route.

      Transit planners who adhere to the “circulators don’t work” religion are stuck in the 19th century where all employment and retail took place in downtown. That is not the reality we face today nor will it ever be the reality, nor should it. The route and land use patter I described is actually MORE efficient and LESS costly to serve with transit than the 19th-century hub-and-spoke city.

      Transit CAN be MORE convenient that the automobile IF transit planners start thinking outside their traditional boxes.

  3. Well, let’s see now, would I like to wait ten minutes for a circulator bus that took me to the train station for a train that takes ten minutes to get downtown, or would I like to wait twenty minutes for a bus that takes 40 minutes to go directly downtown?

    Yup, that’s a real tough question, for sure.

    Seems to me that when you have a bus running in the same direction as a train, there better be a damn good reason for that bus. And the vague idea that it’s “going someplace” just doesn’t cut it, because everyone knows that in Seattle you go downtown and transfer.

    All you need to do is look at a map of the Rainier Valley (albeit a good map) and you can see that the transit is on MLK Wy S, the library and historic community are on Rainier, and Seward Park and the beach are east of both of those. That’s three major threads of community that need to be connected on an East-west axis.

    And it’s certainly not as though we’ve already spent so much money on the south end of town that we just can’t afford to spend any more.

    Yes, that was a clever post by the Aussie, but I’ve heard a duck fart underwater before.

    1. “Everybody knows that in Seattle you go downtown and transfer.”

      In south Seattle you go downtown and transfer. In north Seattle you just go directly there, whether “there” is Ballard, the U-district, Lynnwood, or Kirkland, without ever going downtown. You only go downtown if it’s between you and your destination.

      But if you’re in the south end and you want to go to some other place in the south end, you usually have to go to downtown first.

      1. Even in North Seattle sometimes it makes more sense to transfer downtown rather than attempt a cross-town bus. For example I find it faster to take a 71/72/73 downtown from the U-district then a 15 or 18 to Ballard rather than attempt to ride a 44 directly there.

    2. It really all comes down to frequency. Circulators or any system that requires a transfer is inconvenient and will discourage ridership IF the frequency is down around 30-40 minutes. Get the frequency up to 5 minutes (which is not impossible with short circulators) and transfers cease to be the Achilles’ heal of transit systems and actually can increase user convenience.

  4. Circulator services tend to be operated by vans and farmed out to low-bid non-union companies. The vans conversions cant handle the work and either fail or end up costing alot to maintain, and the contractor has to cut its other costs somewhere else (usually in driver wages).

    And like Brian says, the agency also has to cut traditional services since it is broke

    That’s a lose-lose-lose senario for those of us rallying for hi-quality reliable transit.

    Want a ride to your house? Hail a cab! Government should not be subsidizing away the already limited supply of customers taxi-divers have to solicit from.

  5. Thanks for the link; I’m glad you found my post on Rainier Valley useful. For the record, I’m not an Aussie, just an American based in Sydney. I grew up in Portland and have done transit planning projects for many Northwest agencies, including the City of Seattle Dept of Transportation.

    My blog offers commentary of public transit planning with a bit of a focus on west coast North America, since I still think of that as home.

    Cheers, Jarrett

  6. By the way, when advocating for service design changes in Seattle, be sure to look at a document called the Seattle Transit Plan, completed in 2005 by SDOT. It’s here:

    The plan includes a long-term policy network of high-frequency services, called the Urban Village Transit Network. The plan is adopted City policy, so it could be referenced as a way to attract City support or opposition to particular Metro proposals. For example, the UVTN does advocate a direct east-west service between West Seattle and Rainier Valley.

    Full disclosure: I was the lead planning consultant on the network design aspects of the Urban Village Transit Network, but that doesn’t make it a bad idea!

  7. it is not a faith-based conclusion, but one based on reason and experience. circulators have been implemented many times in many markets and do not attract as many riders per platform hour as alternative service designs, such as those suggested by Jarrett. rather than serve a single purpose, that of taking riders to a rail station, it is more efficient for routes to serve multiple markets or trip purposes; a route may take riders to a rail station and then go on to a market not previously served as well, say First Hill or Georgetown, in the example of southeast Seattle. both buses and service subsidy are scarce and should be used well.

    1. eddiew,

      Did you even read the post? The entire point is that the “double-loop” circulator is more than a way to get to the rail station.

  8. I hate to be one of those NIMBY’s, but I live on Beacon Hill, and Rainier Valley is scary! I do *not* support making it easier to connect from one to the other. I just don’t want the violence spilling over up the hill. It’s surprised me how well the hill keeps much of the riff-raff out, but it does.

  9. A lot of confusion here seems to arise from not distinguishing a true circulator from a bus with a poorly designed route.

    A circulator circulates, like the people-movers at the airport. In the case under discussion, one or two buses would circulate between the lake shore and MLK Wy S. Like the elevator in a building, it does not exist to collect revenue.

    The NYC subway system has two circulators on Manhattan, the 42nd St and the 14th St-Canarsie. I don’t know if the 42nd St goes anywhere, I rode the Canarsie line to the end one day and regretted it. They’re both busy during normal business hours.

    The lack of real circulators is why the downtown free-fare zone is useless to anyone who can still walk. You can walk the entire length of the ‘free’ zone during the time you wait for a bus that is “going someplace” to come and give you a lift.

    A circulator is different from a cross-town bus. As noted above, north of the ship canal there are lots of cross town buses. South of the ship canal they call them circulators and say they won’t work. Plus ca change, plus c’est la meme chose.

    Another type of circulator is like the Health Sciences Express, serving the hospitals, university, and some places where people get off other buses or ferries. This is a fantastically efficient and well patronized service. The distances are larger but they don’t make many stops and they go directly where the riders need to be.

    Of course, the most common form of circulator is the elevator or escalator. When transit planners tell us circulators don’t work, they’re really telling us more about themselves as planners than they are about circulators.

    1. Sound Transit itself is implementing a Tukwila Link-Airport shuttle. It will be interesting to see what will people choose: shuttle to Link or the 194?

      1. Will the shuttle go to the shuttle area of the airport’s access roads, or to the public transit section?

    2. Well, as one serial catowner to another, yes, you can build a circulator into a neighbourhood the way you build an elevator into a building, but its operations do have a cost, and if you want KC Metro to pay for it, the cost has to make sense compared to other things that they might spend the money on.

      The whole point of your downtown free fare zone is that you can use any of the buses running along the avenues to travel within downtown. I agree that there are problems of legibility, especially as compared with Portland’s simpler system, but there are plenty of buses there.

      1. Well, actually, the point of the ‘free fare zone’ is that they avoid the expense and delay of collecting fares downtown. A few people may get a free ride, but it in no way resembles the circulators at the airport, where you go and wait a few minutes, the tram pulls in, you board, and away you go.

        Of course circulators have a cost and Seattle is arguably nowhere near the density that makes paying that price worthwhile- at least, not the parts of town that don’t have elevators.

        In the days of the trolley, the street railway’s motto was ” A Car in Sight at All Times”. That’s a far cry from the present experience of riding a bus in the Rainier Valley. I suppose a few hundred invested in modern phone-computers would solve the problem. I’m so old I remember when spending that amount on a car solved the problem, and at the time it seemed to work pretty well. Death, where is thy sting!

  10. Bellevue wants a circulator that I think is doomed to fail. They made it a short route to increase frequency, but the proposed route doesn’t go anywhere useful. To me that places I want to go but take longer to get to are Old Main and the hospital/whole foods. The circular will make me wait 12-15 mins to get somewhere I could walk to as fast.

Comments are closed.