Metro’s Battery Bus Plan Hinges on Need to Grow Service and Revenue

A battery-powered bus during a press event.
Battery buses like this one are supposed to replace Metro's diesel hybrids within 20 years. (Photo by Ned Ahrens / King County)

In September, King County Metro filed a report on their battery bus implementation plan. Like the Kenmore water taxi report, this paper is the result of a proviso within the 2019-2020 King County budget. The County wanted Metro to discuss major milestones for the 2021-2022 biennium, procurement plans and infrastructure schedules through 2040, cost projections, study and evaluation of battery bus implementation, and a preliminary high-level financing planning in the battery bus plan.

Metro’s battery electric bus (BEB) evaluations began with a 2016 purchase and operation of three short-range BEBs in Bellevue. The short range buses have a 25-mile range and 10-minute charge. The agency’s electric bus program currently exists with 174 trolleybuses, a small fleet of 11 short-range Proterra BEBs—expanded from three—recently concluded testing of 10 long-range but slow-charging BEBs, and an order of 40 long-range (140-mile range) BEBs from New Flyer. Today, 12% of Metro’s fleet is zero-emission buses.

Buses at the Eastgate park-and-ride charging. (Courtesy of King County Metro)

The new buses will be supported by a new South Base Test Facility that will open in January 2022, roughly when the buses will begin operation. Close by, a new interim base at the south campus will be electrified in 2025 and be able to support 105 more battery buses. South King County will host this fleet first as the region is worst effected by air pollution, as discovered by a 2017 study. The study also provided the basis of Metro’s 2040 target for a zero-emission fleet, which the analysis showed would be attainable.

Testing completed in 2020 showed expected and above-expectation performances from the 40-foot buses for range, and a 60-foot New Flyer bus was able to outperform current diesel hybrid buses in snow due to a multi-axle configuration. Cold weather is where performance struggled for the BEBs. Drivers were also happy driving the BEBs. Study on electric base conceptual design found that a “Bridge/Gantry” designed base performed the best overall, and was dominant in future-proofing and efficiency.

The current fleet and fleet to electrify considering a fixed fleet, not including trolley buses (Courtesy of King County Metro)

The implementation plan was developed with all of Metro’s findings in mind, but also with Covid impacts and no additional funding sources in mind. In Metro’s assumed trajectory for the report, service reduction will occur in 2021 exacerbated by expiration of the Seattle Transportation Benefit District—thankfully that’s been renewed and augmented with a $20 car tab fee passed by City Council. Impact from the pandemic will result in service reduction between 2024 and 2027. A countywide transportation ballot measure was supposed to give Metro the resources to grow their fleet to 1,800 zero-emission buses in 2040, pandemic uncertainty pushed the County Council to delay the effort. Worryingly, with Metro’s gloomy outlook its 1,187-bus diesel-hybrid fleet would shrink to 940 battery buses by 2040. That’s a loss of more than 200 buses from today’s fleet. The electric trolley fleet would grow from 174 to 204 for a total of 1,144 Metro coaches in all. Light rail expansion will pick up some of the slack, but buses will still be essential for feeder and local service–not to mention RapidRide expansion plans–so shrinking the fleet hardly seems wise.

A picture of how a gantry/bridge design would be laid out. (Courtesy of King County Metro)

The Timeline

In their implementation plan, Metro outlines two timelines for a completely zero-emission fleet. One finishes conversion by 2040 and the other in 2035, acquisition plans are the same for the two plans up to 2035. The 2040 timeline opts to space out the bulk 2035 order that the 2035 plan leaps for. In addition, the last new 13 diesel-hybrid buses will arrive in 2023 for RapidRide G‘s completion, and Metro plans to expand the trolleybus fleet by 30 buses in 2029.

Simultaneously, Metro would also be constructing capital upgrades to provide charging infrastructure to expand BEB capacity. Here the 2035 implementation plan accelerates the BEB infrastructure timeline to provide capacity for 1,393 battery buses in 2037 rather than 2040. Problematically, this timeline also limits the ability to expand the battery bus fleet between now and 2030, as capacity and projected fleet size closely follow one another until the mid 2030s. Even if the fleet was the same size as Metro’s projected capacity, we wouldn’t be able to even return to today’s fleet size until 2036 in the accelerated 2035 timeline if Metro adheres to a zero-emission fleet.

Battery bus capacity will be constrained until the 2030s as this graph shows.
The 2040 plan graphed out, the solid line represents the battery bus fleet, and the dashed line represents the battery bus capacity. (Courtesy of King County Metro)

Somewhere in the timeline, a more sophisticated charge management system will also be developed and implemented to optimally charge a burgeoning fleet of BEBs and control electricity costs and usage. This will involve a system of chargers, battery packs, and software to minimize charging during peak electricity demand, prioritized buses to require more charging, and automatically lower or stop power levels to buses that do not require more charging. This is aided by the mix of short-range/quick-charge and long-range/slow-charge buses that will help increase the flexibility of the system.

The cost of electrifying the bus fleet

This fleet and infrastructure update will be by no means cheap, two cost scenarios are run to update 2017 results that initially found a 6% increase of Metro’s life-cycle costs for a zero-emission fleet transition. The difference went down to 2% when factoring in societal costs like emissions and noise. These numbers are mirrored in Metro’s favorable BEB scenario, where input variables are adjusted to favor battery buses and the only difference is that BEBs are 1% cheaper than diesel-hybrid buses when considering societal costs. Positives begin to evaporate when Metro runs down the costs in the moderate scenario that the agency is most confident in as an accurate estimate.

In the moderate case, battery buses would cost around $1,916,000,000, or 53% more than hybrid buses, in 2019-2040 lifecycle window, and 42% after societal costs. That respective cost differential translates to 270,000 and 237,000 service hours over the 19-year period. According to the report, “The analysis assumes fueling and charging infrastructure are amortized over the life of the infrastructure. Electrical infrastructure has an assumed asset life of 40 years, direct vehicle charging infrastructure has an assumed asset life equivalent to the vehicle life of 15 years, and diesel underground storage and pumps have an assumed asset life of 40 years. Additionally, the costing is based on maintaining the current diesel-hybrid fueling infrastructure compared to building new BEB charging infrastructure.”

A financial breakdown of acquiring, operating, and disposing of buses and their associated infrastructure shows the projected cost increase that electrification will have.
Costs broken down into categories, details for each can be found the report appendix. (Courtesy of King County Metro)

Discussed in detail is Metro’s confidence that operating costs for battery buses will be greater than hybrids. This goes against typical expectation that electric is cheaper to operate than hybrids. They point to underestimation of infrastructure costs in the 2017 study, as their data has improved. Maintenance costs are acknowledged as volatile and possibly cheaper than hybrids. The agency also specifically addresses a June 2020 study by the National Renewable Energy Lab that found that BEBs would make up the difference with hybrid buses after three years. They note that the results aren’t applicable to Metro as the lab uses retail prices of diesel versus wholesale cost, the modeler’s declining costs compared to expected increasing electricity costs for Metro, Metro’s higher estimates for equipment costs, and the inclusion of grant funds that Metro sees as unreliable.

One of Metro’s short range BEBs (Courtesy of King County Metro)

On grants and funding, the agency has historically funded fleet acquisition from cash and grants. Debt financing, leasing, and even private partnerships are also presented as options. A specific financing plan isn’t present in the plan, but the agency notes that new revenue and various financing methods would be needed to fund the task of transitioning the remaining 89% of the fleet to zero-emission vehicles.

King County Metro ends the report on their proposed 2021-2022 budget to support 260 battery buses by 2028, and their commitment toward a zero-emission future. As nice as that sounds, the agency’s report paints a rather bleak future with a smaller fleet and an expensive transition to a zero-emission fleet. This future only exists in a world where the County and its residents don’t properly support the agency’s public transportation.

Without additional funding, the County will go in the wrong direction in a world where car tires are collapsing local salmon runs, and vehicle miles traveled needs to drop 30% by 2030 to keep global warming under 1.5° Celsius even in the optimistic scenario where the United States adds 50 million electric vehicles by then. Hopefully, a countywide transportation ballot measure is able to materialize to help both electrify and increase the service hours of Metro. Otherwise, I’m not sure if the cost of electrification is worth the service hours we could get with a hybrid fleet.

Clarification: This article has been updated to note the expected fleet of 204 electric trolleys in 2040.

We hope you loved this article. If so, please consider subscribing or donating. The Urbanist is a non-profit that depends on donations from readers like you.

Shaun Kuo is a junior editor at The Urbanist and a recent graduate from the UW's Jackson School. He is a Seattle native that has lived in Wallingford, Northgate, and Lake Forest Park. He enjoys exploring the city by bus and foot.


  1. My first instinct about bus electrification is that it should happen, but it should ideally be paid for with federal grant money, rather than local tax dollars.

    Right now, we’re in kind of a chicken an egg problem where switch to electric buses increases cost. But the cost increase is largely due to the fact that there’s not yet much demand for electric buses, so they don’t have the economies of scale that diesel buses do. Similarly, charging infrastructure for electric buses doesn’t have the economies of scale that diesel storage tanks do.

    For any one specific transit agency, paying the extra cost to be the first user of a new technology is never worth it, diesel buses end up perpetuating indefinitely. Only with some larger entity – such as the federal government – offering subsidies to keep the costs of electric buses down (similar to what they do for car owners with the EV tax credit) – can this chicken-and-egg cycle be broken.

    Another way the feds could help get the economies of scale going is electrifying the vehicles in their own fleets. Electrifying the postal service vehicles would be a good place to start.

  2. Yeah, it is hard to imagine an aggressive approach to electrification as being worth it. Buses are generally a lot better for the environment, even when they are using diesel. I would imagine there are lots of cases where switching over to an electric bus is not the most cost effective way to reduce overall emissions. For the same amount of money you could run the diesel buses more often, and get more cars off the road. I don’t know if the county has bothered to study that aspect of the issue.

    I would guess a gradual transition is the best value (from a transit as well as environmental standpoint). As you noted, a large number of our buses run under wire — and they run a lot.

    Which gets me to another point. Fleet size is all about peak use. It is common to see buses on long term layover, or going back to the bus barn after being used during peak. This makes sense, since frequency on all the routes goes down outside of rush hour, sometimes to zero.

    If we build the infrastructure to run electric buses all day, then it could mean that only our peak buses are diesel. This means that those buses run only part of the day. It also means that they often run on the freeway, where they are relatively efficient. In both cases the buses would emit a relatively small amount of diesel.

    If we live with a smaller fleet, it means less peak service. There are savings to be had as Link expands, and in many cases, it is a straightforward trade: instead of running buses to downtown, you run buses to a Link station, but more often. This would be revenue neutral in terms of service (i. e. driver time) but also use the buses more efficiently (thus reducing the need for so many buses). As you wrote, doing that would not be easy, nor necessarily adequate. It still might require a reduction in service. During rush hour, this basically means more crowding, although if you simply killed off some routes, then people would presumably drive.

    Metro could do a lot more than they are doing now, though. Basically, you connect more buses at Link stations and avoid sending them downtown. Consider the situation after Lynnwood Link (the last major expansion in our system for a while). By then Link will be at Federal Way and East Link will be mostly complete (with only the minor addition to downtown Redmond left). By then the SR 520 project in Montlake would be done as well, which would greatly improve the connection from the UW station to 520 (buses would essentially be in their own lane the whole way). At that point, you could do several things:

    1) No buses on I-5 in Seattle. They would serve a Link station instead.
    2) All 520 buses terminate at the UW, while all I-90 buses terminate at Mercer Island.
    3) More buses terminated at South King County and Sounder stations. A lot of these are Sound Transit buses (I’m not sure how that factors into the system — I don’t know if they count as Metro buses).

    That still leaves a lot of express buses, as well as just buses that run more often during rush hour. The 3/4 for example, could hardly be called an express, but it runs more often during rush hour. Despite the significant savings from truncation (which would enable the same peak ridership with fewer buses) I doubt there would be huge savings. Most of the buses from the south would be the same, as would buses crossing the ship canal west of I-5 (along with buses like the 44, which doesn’t cross the ship canal, and is still peak oriented).

    In any event, the idea that we can get by with a lot less peak service is independent of service levels. We could increase service (substantially) without increasing peak service (especially since buses run fairly frequently during rush-hour). Or we could have low levels of service, along with fewer express buses.

    • I do not believe ST bus ‘count’ in these numbers. Those buses do take up space & resources, so they are relevant when looking at KCM (and CT, PT) bus base capacity and labor capacity, but are generally excluded from the counties’ capital plans and discussions around financial capacity.

    • Good point about fleet size driving peak capacity, not total capacity, and how peak capacity is less important once ST2 is built out. If KCM is running a fixed fleet, that probably means more annual service hours given the likely post-ST2 operating patterns.

      Also, if KCM is keeping the the fleet flat, that removes the need to expand bus base capacity, which would be odd given a new bus base remains one of the largest (unfunded) capital expenditures in the Metro Connects long term plan; dropping the need to fund an entirely new bus base would more than offset higher lifecycle costs of BEB vs Hybrid.

      I think there are too many unknowns to determine if BEB will be more or less cost effective that Hybrids over a full lifecycle. The non-fuel lifecycle costs of BEB is still very hazy, and the cost of diesel fuel a few decades out is a complete unknown. My takeaway from the financial analysis is that the error bars overlap so either options is fine, financially. To me, potentially higher lifecycle costs of BEBs is simply a headwind risk KCM needs to be aware of and contingency plan for; it might impact the pace of electrification, but not the end goal.

  3. Thank you Shaun for this article.

    Your article mentions Metro has issued two updates to its original 2017 formula that calculates that the life-cycle cost of conversion to zero emission buses drops from 6% to 2% based on “societal costs”, “like emissions and noise”. Can you clarify how Metro calculates societal costs, and all the factors that formulate the drop from 6% to 2% in life-cycle costs. I did not quite understand that. Also can you clarify what factors have changed from Metro’s 2017 calculation of the cost to transition to zero emission buses that required two updated cost predictions through 2035 or 2040. I didn’t quite understand why Metro is changing its 2017 cost formulations, and whether the costs increase or decrease for the different options from 2017.

    What I do understand, and have posted about before, is in basic terms the question is whether more Metro service (routes and frequency) are more important than carbon emission savings, both from diesel buses to hybrid buses, and hybrid buses to all electric buses. That is a very difficult question indeed, and forces one to choose between two ideals: transit vs. climate change. Has Metro asked those who actually take the bus, and have to take the bus, their opinion on this difficult choice, and are there any estimates of total carbon emissions between diesel, hybrid and all electric buses through 2035 or 2040? The cuts to transit from going zero emissions are pretty well known, but not the amount of carbon savings.

    You conclude your otherwise excellent article with a link that states “car tires are collapsing local salmon runs”. Don’t buses use tires too, and due to the weight wear those tires faster than a passenger car? Freight trucks too?

    I thought this bit of the article (and only this part) was part hyperbole, because car tires are not collapsing local salmon runs (and really have nothing to do with the main point of the article), and the compounds in tires can easily be changed for buses, trucks and cars if they are proved to be harmful to salmon.

    A more relevant cause of salmon declines are the culverts the state is fixing per court order. According to the Seattle Times article on Sunday, the cost remaining for this fix is $3.5 billion statewide, and the legislature is thinking of moving this massive cost from the transportation budget to the general fund budget to free up transportation spending based on limited revenue, except then salmon will compete with funding for general fund appropriations, from homeless to healthcare to education to equity to to education to all the other general fund expenditures, although there might be more funding for transit.

    Metro is simply doing what the state legislature will have to do for a few years: make choices due to reduced tax revenue. Culverts, transit, climate change, education, health care, bridges, and so on.

    • Sure I can help with those questions. On the changes from the 2017 cost benefit analysis and the more recent cost update, I believe the main cost difference is the cost of electrical charging infrastructure. In 2017 Metro had assumed that charging equipment similar to the kind for light-duty vehicles could be used, they have since found that the industry had moved toward an overhead gantry system that would be more expensive. You can find additional and smaller details on differences in the report itself.

      Similar statement for the societal costs. This refers to emissions and noise, which metro has a dollar value calculation for deep into its appendix for the report. It basically gives the externality of noise and emissions a dollar value, and since battery buses have less of them compared to hybrids the value for battery buses becomes more favorable when those costs are included.

      On the salmon point, my understanding is that buses effectively take a fair amount of cars off the road, and with that in mind then I’m assuming that factor would give a more favorable number on tire emissions per person for buses. We’ll need actual study on this, of course. Brandon wrote their thoughts on the issue, and I agree for the most part, so I decided to link the relevant piece.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.