In 2014, Seattle voters approved a $60 annual vehicle license fee and 0.1% sales tax increase to fund bus service in the city, using the framework of a “Seattle Transportation Benefit District” (STBD). Originally intended to avoid deep service cuts, by election day the revenue picture had improved enough that the package instead improved service on crowded routes. This measure, dedicated to operating funds, is not to be confused with the 2015 “Move Seattle” measure that funded capital improvements in roadways to create more transit priority, among other things. Late last month SDOT released its “Year Two Performance Report“* to promote all the things this money had accomplished.

The measure generates about $50m per year, funding over 300,000 annual service hours, the Youth ORCA program, and greater outreach on low-income ORCA LIFT. Those hours revised the night owl network, and raised the percentage of households a 10-minute walk away from a frequent (< 10 minute headway) bus from 25% to 64%. However, this is a somewhat peculiar definition of 10-minute headways, as SDOT Transit Director Andrew Glass-Hastings explains:

Our definition (for this metric) is an average of at least 10-minute service over a span of 13 hours (6:00am to 7:00pm). .. While some runs may have slightly more than 10 minutes the average is 10 minutes or better, which is a very high level of frequent service.

So the C and D lines, with 12-minute headways mid-day, have enough 6-minute spacing during the peak to allow each to average 6 buses per hour over 13 hours. Hastings added that SDOT could reach its 2025 goal of 72% of households as early as next year, using “about 30,000 hours in the Routes 41, 45, 49, 70 and 120.”

Another part of the STBD was “regional partnership” funds to allow other jurisdictions to combine with Seattle to fund mutually beneficial service. As one might expect, suburban cities have been a little less excited about finding tax dollars for this. However, Mercer Island has partnered with Seattle to fund Route 630. King County Metro has helped fund 11 routes that aren’t sufficiently contained in Seattle to be solely funded by the STBD.

See the report for tons more detail on exactly which routes got what, what routes have improvements still to come, and other statistics (like ridership) that are clearly related, but cannot be directly attributed to STBD improvements.

The STBD concludes in 2020. It will be up to Mayor Durkan’s administration (and voters) to determine what replaces it, if anything.

* Written by many people, including STB alum Adam Parast.

27 Replies to “SDOT Releases 2-Year Report on Bus Service”

  1. The TBD revenues dedicated to more bus hours (300,000 annually) now are ~$45 million. That’s about $150 per service hour. Why is the TBD paying Metro so much — that’s higher than Metro’s cost, and considerably higher than the average bus service/hour shown by the recent NTD data. Sweeet deal for Metro!!!

    2. The funding agreement between the city and Metro present the hourly rates by vehicle type operated. Articulated buses range from $158/hour for RapidRide to $166/hr for trolley buses. 40-ft buses are $137/hr. Metro’s systemwide average for 2016 off its website is $140/hr. The city’s top investments are going into RapidRide and major routes that mostly use articulated buses. The city’s rate also includes compensation for Metro’s management functions which I don’t think is factored into Metro’s own rates.

      1. The NTD average is about $120 per hour — why so much being paid to Metro? There are no additional costs for RR service, if anything the drivers spend less time in their seats working in and out of bus stops than the drivers on milk runs.

      2. The NTD average is not relevent to the question of whether the city is paying Metro more than it costs Metro to operate the service. Glenn posted those costs from around the same time the agreement I posted was reached. They are about the same.

        RapidRide has infrastructure that normal buses don’t have like the offboard ORCA readers and real-time info at stations and the fiber-optic/wireless communications network for TSP. Those have operating costs too.

  2. Fascinating stuff. This brings up a question I’ve had for a while. How do you find route specific information for Metro. Ideally I would like to see stop information for each route, but I can’t even find ridership per route. Anyone know how to find this information?

      2. Thanks. I still couldn’t find any stop level information (maybe they just don’t gather it). I thought in the past they did.

        @Arthur — What is PDR?

  3. Reading through the report, there are a couple of things that make me a bit wary. Hopefully they aren’t driving too many of the decisions when it comes to service and routing. First is trying to reach a somewhat arbitrary goal. Ten minute average service is great, but to achieve that, it looks like they will put money into lines that have better than 15 minute service (but less than 10). That might mean improving a route from 15 minutes midday to 12 minutes. That is nice, but I would much rather see a different route changed from 30 minutes to 15.

    The second is the focus on “access to frequent transit” as if all transit is the same. A bus route that zig-zags around may achieve the objective, but have few fans, because they would prefer a more direct route. The 67 comes to mind, if you are trying to get from Northgate to the UW.

    I guess in both cases I am concerned that they don’t focus too much on achieving an arbitrary, and somewhat abstract goal.

    1. Wait, what’s wrong with the 67? A straight shot down Roosevelt can’t possibly get more direct, unless you’re referring to the hook it does to serve Northgate TC.

    2. He must be referring to either the button hook or Campus Parkway. The button hook was a tradeoff over 5th Ave NE to get more service to denser Roosevelt. Campus Parkway is a major transfer point, essentially an unacknowledged transit center, so it’s essential.

      If you’re concerned about zigzags, the 62 comes to mind. The 45 also makes several turns, but I actually find it useful as extra frequency from 65th & Roosevelt to UW Station, and as direct service from 65th & Roosevelt to the Ave. So this overlap between the 45 and 67 seems to be useful.

    3. Yes, I’m referring to the button hook to serve Northgate. Any time you can get off the bus, walk a few blocks, then catch the bus again (with plenty of time to spare) it isn’t a good sign.

      1. Just think of the route as going from the U-district to 5th Ave. and Northgate Way, with the stretch down 5th into the transit center as existing merely for Metro’s convenience, since the transit center is the obvious layover spot. There is little reason to actually ride the bus to the transit center.

      2. @asdf2 — If that was your goal, then you could just continue on Roosevelt up through Pinehurst and end where the 73 ends. That would be just about as fast, and provide much better service for the folks in Pinehurst (the Roosevelt/15th corridor). At that point, you could probably just kill the 73.

        I really do think that Metro assumes folks will ride the bus all the way around, from Northgate to the U-District. There really is no better alternative. This is fairly typical for Metro routes — they really aren’t concerned about speed. Their motto should be “We will get you there … eventually”. There are dozens of examples, many of which have been mentioned on this blog, but this one I find especially irritating, because it is new. It is great that we have a lot of new, very frequent bus routes. Many of these don’t go downtown, which is a big change. These types of routes can form a great network. But to do that, you want buses that are going straight, not wasting their time looping around.

      1. Better that this happen at the Spring 2018 service change, please. A decade of non-enforcement by SPD has been a decade too long.

  5. 2 questions – in the leftmost graph, does the fact that blue 2017 = 64% while yellow 2020 = 53% mean that ST has met its goal 3 years early? Though their odd definition obscures the fact that transit needs to be frequent all day. This definition essentially boils down to “78 buses between 6am and 7pm” By that logic, the combined 312/522 is almost frequent despite having 30 minute service after 10am.

    What’s with ORCA LIFT? Only 1 out of 5 eligible Seattlites have ORCA LIFT cards? Only 1 out of 14 King County residents outside of Seattle? That seems very low. Is it a combination of other people not wanting to ride transit and not knowing about the LIFT program? Does LIFT have a surprisingly difficult requirements to sign up?

    1. The definition of frequent seems to work. Theoretically, there are buses that are very heavily commute oriented that could pass under the bar, even if service is infrequent during the day. But as it turns out, there aren’t any. The 41 doesn’t even make the cut, despite the fact that it runs every 15 minutes during the day, and a lot more frequent than that during rush hour.

      >> the combined 312/522 is almost frequent despite having 30 minute service after 10am.

      I guess it depends on what you mean by “almost’. By my count, there are 60 trips in that period (give or take). That means you need 18 more trips to make this “frequent”. Again, theoretically you could cram those into the rush hour period, but why would you? Ridership is higher during rush hour, but ridership is decent during the middle of the day. Add enough service, and you might see an increase in ridership *per bus* as folks switch to the slower (but currently way more frequent) Link/372 combination. Add 2 buses per hour in the nine hour period (10:00 AM to 7:00 PM) and you’ve achieved your goal. This means 15 minute service between 6 AM and 7 PM, with way more frequent service during rush hour. This would be an outlier, as the only bus that doesn’t have at least 12 minute headways during the day.

      But even then, this combination wouldn’t pass the mark, because it isn’t one bus route. They are remarkably similar, but the 522 goes out to Woodinville, while the 312 does not. For many trips, they are the same, but if you are going to combine bus routes in that manner, you might as well add the 372 in there. If you want to go from Lake City to Kenmore, you have very frequent service. The 372 runs every 15 minutes (all day) while the 522 runs every half hour. If those six buses per hour were synchronized, it would achieve the goal — by the strictest of standards — for this segment, since it would be every ten minutes during the day.

      But SDOT has decided not to combine segments in this way, because buses like this aren’t usually synchronized. By focusing on individual bus routes, they have managed to come up with a pretty strict standard that seems to work fairly well. If you look at the chart of buses that meet this standard, as well as buses that come close (page 51), you can see that the 312/522 is nowhere to be seen, while the 372 is on the list of those that are close.

      1. My main argument was that this definition seems biased in favor of rush hour commuter oriented buses. Rather than a more common sense definition of buses that are frequent all day.

        I stand by combining the 312 and 522. Other than the Bothell-Woodinville tail and a few local-only stops, they are essentially the same route. At least I’ve always viewed it just as Metro supplementing a ST route during rush hour. I counted ~65 trips. Which means that if there was 1 additional trip/hour over the whole 13 hour period, it would hit the threshold despite only having 20 minute service during daytime, which is hardly frequent.

        But if your argument is that the routes need to be identical, than the 41 doesn’t even come close to being frequent. Once you throw out the 20+ trips that terminate at 125th St, there’s nowhere close to 10 minute service.

      2. My main argument was that this definition seems biased in favor of rush hour commuter oriented buses. Rather than a more common sense definition of buses that are frequent all day.

        Yet you can’t find a bus that actually fits that criteria. That’s my point.

        Look, I had the same reaction. I’m sure Martin did, too (which is why he mentioned it). But when the dust settled, and you actually look at the buses that achieve that “frequent service” mark, as well as those that come close, we can’t find a single bus route that is biased in that fashion. Let me copy it for you. The buses that meet the threshold are:

        C, D, E, 7, 3/4, 36, 44, 48

        All of those are frequent in the middle of the day. Some of them very frequent.

        Now look at the ones that are close:

        5, 8, 10, 12, 2/13, 21, 31/32, 40, 41, 45, 62, 65, 67, 70, 75, 106, 120, 372

        Again, all of those bus routes run every 15 minutes (so far as I know). So basically, there are over 25 cases where the criteria works just fine, and not a single outlier, despite our initial fears. I don’t see the point of quibbling over the definition, when the definition seems to be working quite well.

    2. What matters is MINIMUM FREQUENCY!!! That’s what’s great about RapidRide: it’s guaranteed 15 minutes until 10pm every day. That makes it easier to visit West Seattle or Aurora or Ballard any time without worrying that you might have to wait or won’t have enough time. I’ve been going to West Seattle more for recreation because of it (Lincoln Park, Alki, California Ave businesses); in fact it’s become my largest recreation destination. And if I lived there I’d be glad I don’t have to keep my trips to weekday and Saturday daytime to avoid the 30-minute periods, as I used to do with the 5, 10, 15 and 358, and now do with the 11. I’m starting to get bummed about Metro’s long-range plan because I thought “Frequent” meant RapidRide frequency (15 minutes until 10pm every day) but now I’m hearing it may only be weekday daytime. Which means that those lines marked “Frequent” may get no more service than today. Sigh. The reason New York and London have such a low percentage of cars is that transit is frequent full time, so you just step out to the curb anytime and a bus or subway comes in 3 minutes or 10 minutes (or night owl 30 minutes). Our evening and Sunday service shouldn’t be Chicago’s and San Francisco’s night owl!

      I understand that Metro has limited funds and all that, but full-time frequency should be a clearly stated goal, with a phased plan to get to it.

      Look at all this uncertainty about 405 BRT. Link has stuck to its promise of 10 minutes until 10pm, so I trust that all the other lines will be like that, and hopefully the BRT lines, but ST hasn’t articulated a commitment, so that means maybe it will and maybe it won’t.

      1. One thing that’s bothered me is that with all the uptick in tax revenues post recession, the Sound Transit service that the extra revenue has bought has been almost entirely rush hour service, while some of the 2009-era cuts to weekend service have still not been restored, with no indication from Sound Transit that they will ever get restored (at least until Link finally replaces the bus route).

        In fact, Sound Transit has always published an “unfunded” ST express wish list, which does, actually, include a substantial increase in evening/weekend service, but when push comes to shove, they never find ways to actually fund it. Even ST3, with its huge new revenue stream, is apparently not adding one single trip to ST express service.

        Part of the problem is that as the economy booms, traffic gets worse, so a large portion of the extra transit revenue gets squandered by extra padding that needs to be added to the schedule of every single existing rush hour trip. And, many of the peak-hour routes are, in fact, overcrowded (although, some are not).

        While they have kept their promise about operating Link frequently all day, as long as the ST board feels that their primary mission is rush hour, I’m personally very skeptical that they’re going to demonstrate the same level of commitment to a bus, especially a bus that doesn’t serve downtown Seattle. Even the Issaquah->Kirkland rail line, I’m skeptical whether the weekend frequency will really be better than every 30 minutes. I guess we’ll just have to see.

      2. It’s the low-hanging fruit and the biggest transit need. How can ST turn away when people are lining up to take it, and go e them a bus at a different time when few people ride. Every politician’s statement says we need transit so people can get to work, because that’s what enables them to stay alive and pay taxes and keep the economy running. I see the fifty people lined up at rush hour for the 535. Throwing extra buses at it is just a coping mechanism until Link is built out. If we really wanted to solve the problem right now, we would have approved Forward Thrust or started ST1/2/3 10-15 years earlier.

  6. I cannot find the tax revenue collected for this levy. For the reporting period, $24.7M was forecasted to be expended for the service (page 10). That doesn’t account for costs of service revenue hours vs non-service expense items. How much revenue is collected beyond the $24.7M and what general ledger account is that amount posted to?

Comments are closed.