Fixing US Transit Requires Service, Not Just Infrastructure

TransitCenter has a new video and article with some powerful images saying what I say all the time:  If you want to transform public transit for the better in the US, there’s useful infrastructure you could build, but the quickest and most effective thing you could do is just run a lot more buses.

(Remember, US activists: Don’t just envy Europe; start by envying Canada.  The average Canadian city has higher ridership than the most comparable US city, not because they have nicer infrastructure or vastly better land use, but because they just run more transit.)

TransitCenter’s work uses access analysis to show what’s really at stake.  Increasing bus service by 40% (an aspirational number that still wouldn’t match many Canadian peers) would massively expand where people could go, and thus what they could do.

For example, here’s how 40% more service would expand where someone could get to from a particular point in metro Atlanta.  (The concentric colors mean where you could reach in 10, 20, 30, or 45 minutes, counting the walk, the wait, and the ride.)

Source: TransitCenter (graphic by Remix)

Source: TransitCenter (graphic by Remix)

With a 40% increase in service someone in this location can reach ten times the number of jobs in 45 minutes.  (These analyses use jobs because we have the data, but this means a comparable growth in the opportunities for all kinds of other trips: shopping, errand, social, and so on. )  I would argue that someone at this location would be 10 times as free, because they would have 10 times more options to do anything that requires leaving home.

The transportation chatter in the new administration is about infrastructure, partly because there’s lots of private money to be made on building things, and because building things is exciting.  But if you want to expand the possibility of people’s lives, and seriously address transport injustices that can be measured by this tool, don’t just fund infrastructure, fund operations.  Just run more buses!



19 Responses to Fixing US Transit Requires Service, Not Just Infrastructure

  1. Jonathon January 28, 2021 at 4:30 am #

    As a Canadian living in the US (Los Angeles, now Detroit), it feels like an uncanny valley much of the time. Ask anyone from Toronto, and they’ll give you a litany of complaints about the TTC, but here are some things you would never worry about- missing the bus and having to wait an hour for the next one (every bus route runs every 30 mins or better), getting stranded late at night (every Torontonian is within 30 minutes’ walk of a 24-hour bus route), checking the schedule to see if the train is coming soon (subways run every 5 minutes or less all day), wondering if you should walk to the express stop because you just missed the local (locals always run every 10 minutes or less if there’s a parallel express service).

    In the simplest terms, Canadians expect that their bus will always run more often this year than it did last year.

    • Sean Gillis January 28, 2021 at 5:01 pm #

      Jarrett: “America: Don’t just envy Europe, start by envying Canda.”
      Canada: “Don’t envy Europe, just gloat how much better at transit than we are in the States”.

      Canada really does need to envy Europe – they are way ahead of us.

    • Crispin Cast-Nine January 31, 2021 at 10:12 am #

      In San Francisco I was once told by a bus driver to my face that he was going to skip the scheduled run I had hustled to make because he “didn’t get a lot of breaks” and that I was welcome to wait for the next one… in 30 minutes. So instead I walked for thirty minutes uphill to get home. Another time on the same route, the only reason the driver grudgingly made his scheduled trip was that an older woman who was waiting with me managed to shame him into it.

      That, by the way, is one of the best cities for transit in America — yet I found transit to be so unreliable there that it was basically useless if you actually needed to be anywhere by a specific time. Even if you used one of the so-called “high frequency” routes (don’t even get me started on Muni Metro), you were always running about a one in three risk of being stranded somewhere with a 30+ minute wait.

      • Mike February 3, 2021 at 11:49 am #

        San Francisco is probably the only city with bus bunching even on uncrowded streets. In the 80s and 90s a three-bus route would be scheduled with even headways between the buses, but instead the buses traveled together and arrived at the same time. I’m told this was because of union excesses: the drivers could get away with it so they saw no reason why they shouldn’t travel together. Never mind the passengers who had to wait three times as long or wouldn’t take a bus because of it.

    • James February 1, 2021 at 3:45 pm #

      It’s a shame, many systems at very least have the skeleton of a “reliable” transit system (minus the tendency to route everything into the city center). But so many systems, like my hometown here in Kansas City have such laughable frequencies and service spans, no one outside of the urban core would consider using it, heck even the urban core barley wants to use transit.

      Heck if the US systems were even half as reliable as Canada’s we’d be in a much better place. But sadly most politicians won’t fight for it, and most residents don’t care. Sad and shameful state of affairs.

  2. RossB January 29, 2021 at 12:16 pm #

    Yes! I completely agree. Seattle was one of the few cities in the last few years that saw increasing transit ridership. The biggest change was simply adding more service. The buses run more often, and people use them more.

    There are other advantages. First, it is hard to screw up. In contrast, American capital transit spending is often too expensive or poorly designed (or both). You really can’t screw up when you run the buses more often. We know, for a fact, that it will increase ridership, while making life better.

    Second, the money goes to middle class jobs, not into the pockets of large companies (this is the private money Jarrett mentioned).

    Third, the benefit is widespread. There is no way a city like, say, Akron, could spend a bunch of money on transit infrastructure and get something worth it. But they could sure benefit from running the buses more often.

    This is by far the best approach for the federal government to take.

  3. James February 1, 2021 at 3:45 pm #

    It’s a shame, many systems at very least have the skeleton of a “reliable” transit system (minus the tendency to route everything into the city center). But so many systems, like my hometown here in Kansas City have such laughable frequencies and service spans, no one outside of the urban core would consider using it, heck even the urban core barley wants to use transit.

    Heck if the US systems were even half as reliable as Canada’s we’d be in a much better place. But sadly most politicians won’t fight for it, and most residents don’t care. Sad and shameful state of affairs.

  4. Sahil February 1, 2021 at 3:53 pm #

    Europe here, but USA has a lot to learn from India too. The last mile connectivity of Indian cities, especially Mumbai is unmatched. Trains run every 3 min or less. Even then, they are jam packed. Construction is on for doubling the capacity and a whole new subway system is coming up. Buses don’t really need a time table, they just show up every 5 min. Inter-state train connectivity is great as well, thanks to Indian Railways. Talking about last mile, taxis and rickshaws are very, very easy to find and pretty cheap. Basically, India is probably one of the very few countries in the world where you can spend your life without ever owning a car- rural or urban. Same probably cannot be said about rural Canada. For a country of its size and population, India has excellent transit, and something USA and Canada should learn from. After all India is not only about cows and snake charmers.

  5. STrRedWolf February 2, 2021 at 5:15 am #

    Let us pull back and pull some more context in place, and set the time frame to before the pandemic.

    Transit was being cut because budgets were getting cut. Frequencies going down because the money wasn’t there. Fares going up, sometimes reluctantly (MTA Maryland was forced by the state legislature to raise fares). The bus maps getting redrawn more and more in order to run less and less.

    So yes, run more service. But how are you going to fund it? “The feds will” isn’t an answer, because the feds were cutting back on the funding. It’s not reliable.

    You have to tax locally and dedicate funds. And you have to set it right, future-proof it, so you don’t have to muck with it constantly. There’s no “one size fits all” solution in this case.

    On the flip side, if you run more bus service, and it’s still full… you’re going to run out of road. You need to run faster service that is off-road. You need to lay rail — building subway and commuter rail service. (Light rail, or street-running, is all good on the outskirts but you’re still mixing with traffic in the city and slowing it down)

    Baltimore is that way. Buses were full. Running more service helped but then that got full. Rail was needed… but politics got in the way.

    • RossB February 9, 2021 at 10:03 am #

      “The Feds should” is the answer. The point is, rather than spending trillions on a “Green New Deal”, we should run the buses more often. Very few places have reached the point where rail makes sense. As you wrote, even those places often screw up, because local politics get in the way (they build the wrong thing). Given the inability of Americans to build big rail projects in a cost effective manner (compared to Europe, Japan or South Korea), it would make more sense to to focus on running the buses more often.

      Even in Baltimore they don’t run the buses often enough. Correct me if I’m wrong, but the “City Link” buses are supposed to be frequent, yet some of them only run 15 to 30 minutes ( That is terrible. Obviously there is room for improvement when it comes to frequency. If the feds are going to help when it comes to big capital projects (which they do) then they should help with something far more cost effective: service.

      • Chris M February 21, 2021 at 2:04 pm #

        Better public transit is an integral part of “green new deal” and similar plans around the world. And part of program of every left-green party.

  6. el_slapper February 4, 2021 at 4:15 am #

    As a spoiled European, I have to boast about my new mayor in Montpellier, Michael Delafosse :

    A bid for 77 new tramways for the town, 22 for the future line 5 (still a few years before the works are finished), 30 for replacing the aging trams of the line 1… and therefore 25 to add frequency where needed (probably lines 1 & 2, though it’s not written).

    Those last 25 are exactly what Jarret is speaking about (if I understood well) when he speaks about improving the service. Lines 1 & 2 are overcrowded on rush hours (1 especially around the hospital/university area, 2 towards the quickly-developping south) and really deserve better frequency.

  7. Chris February 8, 2021 at 10:02 am #

    Unless you are a large city with a high peak to base ratio (like Seattle, Los Angeles, Chicago, San Francisco, etc.) additional service, even at off peak hours, will require additional buses. Actually, even large cities like Phoenix generally operate the same level of service the whole day. If a particular route network operates every 30 minutes all day and requires 100 buses, making this route network operate every 15 minutes, even if only in peak hours, will require 200 buses. In addition to the bus acquisition such an increase in the number of buses could easily result in a need to create a new bus garage, or at least an expansion of the existing one. So yes, I agree that increasing operating funding would be effective but an easy “Seattle” fix is only available in a handful of cities, all of which already have excellent transit service.

    Of course, not even discussed is the logistical difficulty of hiring thousands more bus operators. In a time of high unemployment it’s probably doable, if you devote time to trying to reach non-“traditional” bus drivers. However, in times of low unemployment you might find that the kind of people you want driving a bus – good driving record, no drug use, reliable – can get other jobs that are less work and pay better.

    • Jonathon February 9, 2021 at 4:06 am #

      I would say yes, but to this, because there is a flip side to this argument. It’s easy to think that the only way to increase transit ridership is to be able to afford to run buses every few minutes everywhere, all the time, and yes that is the goal, but there are waystations. An instructive example to consider:

      About 10 years ago, Toronto began what was called the ‘Ridership Growth Strategy’. It consisted of an a few planks. The first was planning for the service levels the city *wanted* but couldn’t implement in the near-term, because they didn’t have enough buses for any additional peak service. The second was implementing all of the planned off-peak service increases with the buses they did have. The third was creating a network of main routes running every ten minutes or less all day every day, mostly with small increases late at night (from routes that used to run every 12 or 15 after 10pm, for example).

      The fourth, and possibly most important, was instituting minimum service levels: every route, everywhere, would run at least every 30 minutes every day until 1am, and no resident would be more than 30 minutes walk from a 24-hour bus route (that would also run every 30 minutes or less). And here’s the rub: when a new mayor came into power who was far less enthusiastic about bus service, he demanded the transit commission to cut all underused service hours, aiming to pare back most of these RGS-era improvements. As it turned out, almost all of the new service- especially on those ‘low-frequency’ routes running every half hour all day every day- were actually being used, and cleared the bar avoiding significant service cuts. So maybe your city doesn’t have enough buses to run service every few minutes, but I bet the routes run every 30 or less at rush hours. As it turns out, even the promise of a bus every 30 minutes all day every day is enough to be useful.

      And one last note- in Toronto bus drivers are part of the powerful auto workers’ union, making living wages (~$30/hour) with decent benefits, pensions, and all the good stuff that auto factory jobs used to have. It’s a respectable profession, because even though not everyone is a frequent transit rider, their kids probably are, and they know that the bus will be there for them when they need it.

    • Jonathon February 9, 2021 at 4:53 am #

      Case and point: London, Ontario has a metro population of just under 500,000. Their strategic plan for 2021 has one major item: eliminating all 60-minute frequencies across the system. This is the kind of low-hanging fruit that a lot of American cities could start with.

    • RossB February 9, 2021 at 10:23 am #

      “So yes, I agree that increasing operating funding would be effective but an easy “Seattle” fix is only available in a handful of cities, all of which already have excellent transit service.”

      What cities are those, with the “excellent transit service”? I’m from Seattle, and it is nowhere near excellent. It is nowhere near as good as our closest neighbor, Vancouver. Way more people ride the buses up there, and for good reason — they come a lot more often (and form a grid).

      Your main point, though, is worth considering. Just about all systems have a peak-oriented bias, but some more than others. It doesn’t really make that much difference though. It is just an implementation detail. If you were to double the operations budget of every agency, some would shift money towards buying more buses, while others just hire more drivers. Those that are more peak oriented would get a better value out of the money, but they would still have to spend more money on new buses eventually (to replace ones that would now be used a lot more often). The overall point is to focus on service, not new construction. The fact that it might lead to a lot more bus purchases isn’t necessarily a bad thing, especially if they are made in the U. S.

    • Jack February 10, 2021 at 7:52 am #

      Excellent point on the difficulty of hiring CDL operators. We can’t maintain staff on our existing system. Raising wages could help, but all our local supporters are already stressed by what they perceive as big pay raises for CDL operators when their budgets aren’t growing. As a smaller system, we are also at the mercy of commercial insurance carriers, which refuse to cover us if we train new folks to get CDLs, so we are limited to hiring people who already have them. This has been our biggest stressor for years.

  8. Steve Dunham February 12, 2021 at 9:50 am #

    What’s also important is the effect that low frequency has on transfers. Lately I’ve made several in-person visits to my credit union. The nearest office I can go to is in a suburb about five miles away. The trip takes over an hour because I have to take two buses that run, at best, every half hour, and the transfers between them take 20 to 30 minutes. Walking would be almost as fast overall, but in places the multilane roads have no sidewalks.

    The difficulty of transferring between low-frequency bus routes surely depresses ridership, and the discomfort of the transfers doesn’t help. There are plastic shelters at the Metro station where I transfer on these trips, and the shelters are some help, and there’s a port-a-potty that is padlocked.

    The buses often display the wrong destination sign, and the electronic signs in the shelters display information that often doesn’t match bus arrivals and departures. One said something like “s8:33.” I asked a bus driver what it meant, and he said he didn’t know.

    Unless you have a pulse system, low frequencies can doom bus riders to transfers that are often long and uncomfortable. (And I know that pulse systems are suitable only in certain locations.)

    • RossB February 14, 2021 at 2:45 pm #

      Yes, absolutely. Transit scales, and the reverse is true. If the buses are so infrequent that transfers are horrible, then this pushes the agency to have fewer transfers. This makes the system less effective, and less frequent. Ridership suffers, which puts more pressure on the system. That is the cycle many agencies find themselves in.

      In contrast, if you increase frequency, then there comes a point where the transfer is harmless. This allows the agency to provide more of a grid, which in turn reduces the time it takes to get to a lot of destinations. Increasing service can make all the difference in the world.