King County Metro has a revenue crisis. They are currently facing a $75 million annual shortfall, and without a new source of revenue, they will be forced to institute a 17% cut in service hours.

Most of Metro’s money comes from taxes and fares. So if Metro needs new revenue, then it’s only logical to look to these sources. King County is actively working on a plan to raise the sales tax and the vehicle license fee (paid annually on vehicle registration renewal) to address this shortfall, as well as providing some additional money for local road maintenance. I strongly support their efforts, and will enthusiastically vote for the measure when it hits my ballot.

However, there’s something that’s always bugged me about Metro’s fare structure. If Alice rides the bus for two stops, and Bob rides for 15 miles, they both pay the same $2.25 fare. (And if Bob has a monthly pass, he may end up paying less than Alice on a per-trip basis.) No other transportation service works like this. Taxi fares are based on distance and time. Toll roads cost more the further you drive. Airlines set fares based on costs and demand. Long-distance trains and buses, commuter rail… the list goes on. Only local transit networks have flat fares.

I’m not the only person who finds this worrying. Steven Farber, a researcher at the University of Utah, has found that flat fares seem to negatively impact low-income populations, who tend to make shorter trips than people with higher incomes. His research was conducted on behalf of the Utah Transit Authority (UTA), which is exploring a move to distance-based pricing, as a way of improving equity and financial stability.

Does the UTA have the right idea? Should Metro be looking at implementing distance-based pricing?

In 2013, King County Metro spent $640 million to operate its bus network,while Metro’s fare revenue was about $146 million. In addition, we know that Metro is facing a $75 million shortfall. Therefore, a distance-based fare system would need to raise about $221 million to solve Metro’s revenue problem.

In 2001 (the most recent data I can find), Metro provided 547 million passenger-miles of service. Let’s assume that this increased 10% since then, and so Metro currently provides 600 million passenger-miles of service. This yields a pretty straightforward figure: Metro must collect about 37 cents per passenger-mile.

The first complication is that a pure distance-based scheme neglects the fixed costs associated with picking up and dropping off a rider. Every time the bus stops, the vehicle is delayed a bit. The bus must slow down, exit traffic, open the door, wait for passengers to exit and enter, then finally reenter traffic and resume speed. As an extreme example, imagine a full bus, where every single rider exits, and a new set of riders reenters, at every single stop. It’s for this reason that taxi meters have a “drop”, and that a flight from Boston to New York isn’t half the price of a flight from Boston to Washington. So it would make sense for a distance-based fare system to have a minimum, too.

Suppose that we set a minimum fare of $1, which includes the first mile. The average end-to-end trip length (again, from 2001) is 8 miles. If an 8-mile trip needs to bring in about $3 in revenue, and if the first mile costs $1, then the remaining miles can cost about 28 cents each. We’ll round off and call it a quarter.

Here’s a sample of what some one-way trips would cost:

  • Downtown (3rd and Pike) to Southcenter Mall: $3.75
  • Downtown to Harborview Medical Center: $1
  • Downtown to the University District (45th/University Way): $1.75
  • Ballard to West Seattle: $3.50
  • U-District to Bellevue: $3.25
  • Northgate to Downtown: $2.75
  • Downtown to Black Diamond: $9

There are some implementation challenges with distance-based fares. The main issue is that you need a way of knowing where people end their journey. The simplest way to implement this would be by installing rear-door ORCA readers. Riders would tap their card both when they enter and exit the bus. In 2010, Metro estimated that installing rear-door ORCA readers on all buses would cost $5.5 million. There are other costs, such as training drivers, updating the ORCA programming, etc. (Luckily, we know that ORCA supports distance-based fares, since it is used this way for Link.) In total, let’s assume that this project would cost about $10 million in today’s dollars.

Still, if this system would allow Metro to eliminate its operating deficit, while simultaneously improving equity, it seems like something that we should consider.

This post originally appeared on Medium.

197 Replies to “Financially Sustainable Transit”

  1. NO, NO, NO, a thousand times NO.
    You advocate for a distance based system, yet fail to acknowledge that Metro’s costs are mostly based on a time system.
    Drivers are paid by the hour, not by the mile.
    Buses cost x per operating hour (~$130 last time I looked)
    Overhead is mostly salary driven, also paid by the hour.
    Buses get so many mpg, but consistently burn x gal/hour.
    About the only thing distant based are maintenance supplies like tires.
    A city bus travels much slower than an urban bus, yet many rides take nearly the same time to reach their destination. Since the cost of that trip is about the same, the fare should be about the same.
    As riders only pay for 1/4 of the cost, the difference between a 30 minute trip and an hour trip is pretty small from the riders perspective, and not worth the delay and confusion at the farebox by making the fare collection system complicated.
    VOTE FOR FLAT FARES!

    1. What you’re really advocating for is not a flat fare, but a time-based fare. I could picture it, but it might cause some hostility in the riders when you get a slow bus driver that chats with every rider.

      1. NO Matt, I’m not.
        This bean counter mentality to collect 1/4 of the operating cost is maddening. The other 3/4 of the cost are just turned over to JoeTaxSucker as the cost of doing business with little regard for who pays, how much, and for what logic.
        I’d like one Seattle politician stand up and say, I’m charging everyone in Black Diamond 3 times more to ride the bus, because you are stupid to live so far from the center of Seattle. Shame on you, and I’m making you pay for a poor choice on your part.

      2. So would I, mic. Maybe with a bit less of the “stupid” and “shame” bits. There’s nothing wrong with the decision to live in the middle of nowhere, and many people value that lifestyle. But it costs real money out of a dwindling transportation fund to run buses there, and people should pay for expensive choices rather than spreading costs among others.

    2. But the level of service all these costs are structured to meet is defined by how many miles my butt sits on a bus, eh? If everyone was going only one stop, we certainly wouldn’t have any articulated busses.

      For what it’s worth, the ruthelessly rational and economicly prudent Dutch use a distance-based approach for transit here in my other home.

    3. What you’re saying about time-based costs is absolutely true. But think about it from the perspective of a rider. Let’s say I want to travel four blocks, but because of traffic, it ends up taking half an hour. Should I pay more because I had the privilege of traveling at a snail’s pace? The customer should never be asked to pay more as the quality of service declines.

      Also, you could argue that the real burden is on Metro and their routing choices. If Metro chooses to run your neighborhood bus down a slow local street instead of a fast arterial, it makes no sense to say that riders should pay more for the slower bus. As another example, let’s say that I’m trying to get from Capitol Hill to the International District. Driving down Broadway, this trip would take only a few minutes. On the bus, you’d have to go downtown first, and that would take half an hour. Should the passenger really be asked to pay extra because of how Metro chooses to structure their network?

      1. But your proposal does ask this of riders.

        $3.50 for an hour-long crawl from West Seattle to Ballard, thanks to detours and the downtown penalty and the total disinterest in the “R” in BRT? While those lucky enough to begin or end downtown continue to enjoy their heavily-subsidized and penalty-free peak trippers?

        I can’t think of a worse incentive to reinforce the downtown-centricity of the powers that be, and the screwing of everyone else.

      2. No, Aleks’ proposal specifically doesn’t do that. A fare charged by distance wouldn’t give any discount to a faster peak tripper. In the comments, he clarifies that he wants to charge based on as-the-crow-flies distance, so you wouldn’t even get charged more for the minor geographic deviation to Queen Anne.

        In fact, though it wouldn’t be applicable in your case, charging based on crow-flies distance would make some freeway expresses even more expensive – consider the 355 or 301, for instance, which deviate to I-5 while the local 5 or 358 continues straight south.

      3. “The customer should never be asked to pay more as the quality of service declines.”

        I totally agree with this. We should be incentivizing Metro to invest in capital projects to improve its quality of service.

        We should also, of course, make sure that those who have the most difficulty affording fares have lower fares, and the LIF program as announced is a start. We need to do a lot more, too.

      4. d.p.,

        As other commenters have said, the proposal (albeit with a necessary clarification in the comment threads :P) is to charge based on crow-flies distance, specifically to avoid penalizing riders when Metro makes questionable routing decisions.

      5. @d.p. Which is why it isn’t just the distance that is taken into account. The type of the route must also be considered. Local routes where people are not expected to travel the entire length journey should cost less.

      6. That doesn’t change the fact that you would charge double for the many intra-urban trips that are perpetual teeth-pulling nightmares — that practically dare you to drive next time — while charging next to nothing for Matt’s 29 express to Queen Anne.

        Express buses should see a surcharge. Suburban express buses should see a greater surcharge. And some differential fare should probably exist in the off-peak to rectify the absurdity of charging the same fare to First Hill as to Duvall.

        But a complicated distance-based algorithm that can’t help but further penalize lots of in-city trips, on a transit system that already fails to optimize for urban usage patterns and urban needs, strikes me as unconscionable.

      7. “distance-based algorithm that can’t help but further penalize lots of in-city trips” Can you describe some of these? Most in-city trips seem to be fairly short compared to suburban commuter trips, especially as the crow flies.

      8. But a complicated distance-based algorithm that can’t help but further penalize lots of in-city trips, on a transit system that already fails to optimize for urban usage patterns and urban needs, strikes me as unconscionable.

        What we’re trying to explain to you is that this proposal doesn’t do that at all. Unless your complaint is just that riding the 29 to Queen Anne would cost no more than riding the 2 there, which is less than riding (say) the 2+49 to Capitol Hill? In that case, while that trip definitely does need to be improved, I’m not sure your complaint has merit, because it still is a longer trip distance-wise.

      9. I’m curious to hear some examples of trips that you feel would cost too much in this proposal. By construction, you have to travel at least 7 miles off-peak, or 9 miles during peak, before the fare is more expensive than it is today. Do you really think that a fare increase for 10-mile trips — especially coupled with a fare *decrease* for the spontaneous 2-mile and 3-mile trips that are key to urban living — is a disaster?

        As a few more examples:

        – A trip from Ballard to Capitol Hill would cost $2.25 — the same as today, or cheaper during peak.
        – Ballard to the U-District would cost $1.75.
        – Ballard to Fremont would cost $1.25.

      10. Greenwood.

        There’s your example of a reasonably urbanized and unquestionably in-city neighborhood that suffers a painfully slow, unreliable core service with terrible night/weekend frequency and bad connections to other services — just an overall mediocrity of transit — and that would see its fares rise under this plan.

        Just heading downtown would get pricier (for no gain). But why in the hell would any Greenwooder suffer the indignity of a transit trip to the Central District or to Georgetown, when they’ll be charged $3-$4 for the timesuck and the indignity?

        Remember that the 99 tunnel toll will only be $1.

      11. …And that, of course, presumes that you’d be able to lower the base fare as much as you’d like, which I strongly doubt.

      12. Check your map again, d.p. – I just did, and Greenwood is only 4 miles from the Central District and 6 from Georgetown. Even rounding up, Georgetown would still be the same price, and the Central District will be somewhat cheaper.

        (Your point about the 99 tunnel being poorly positioned and too-cheaply priced still stands, though.)

      13. You’re wrong.

        As the crow flies, 85th and Greenwood is just over 6 miles from downtown, 6-8 miles from various parts of the C.D., and 11 miles from Georgetown.

        If Aleks’ crow-path calculation of $3.50 from Ballard to W.S. is accurate, then you’re looking at similar or higher fares from central Greenwood to most places.

      14. I actually think Greenwood is a great example, because it perfectly illustrates the kind of short trips that people actually make.

        Here’s a map of the top 100 origin and destination pairs for non-work trips in Seattle, via all modes.

        From Greenwood, here are all of the destination neighborhoods:

        – Broadview
        – Bitter Lake
        – Lake City
        – Northgate
        – U-District/U-Village
        – Green Lake
        – Fremont
        – Phinney Ridge
        – Ballard
        – Sunset Hill
        – Crown Hill

        That’s a lot of neighborhoods. Did you notice what’s not on the list?

        – Anything south of the ship canal

        When you look at where people actually want to go, they tend to want to travel to places that are relatively close by.

        The thing is, you wouldn’t know this from looking at current bus ridership. Right now, the 5 is an acceptable way to get between Greenwood and downtown at an acceptable price. But you’re paying the same price to get from Greenwood to Phinney Ridge — and so comparatively fewer people will take the bus for that trip. The vast majority of these trips are being made by car, because who would pay $2.25 for a 30-minute, 2-mile bus trip?

        The outcome is similar (though inverted) to the outcome that you’ve described on RapidRide D, where Ballard-to-downtown ridership is artificially depressed by the LQA diversion, making LQA ridership appear more important than it actually is.

        In practice, this change *would* affect bus ridership. The average trip length would become shorter, as people shift towards the bus for relatively shorter (2-mile to 4-mile) trips, and possibly shift away from the bus for relatively longer trips (8-mile and more). But given that there’s much more *total* demand for the former kind of trip, I’m not sure this is a bad thing.

      15. With costs being proportional to time and revenues proportional to distance, it may become evident to Metro that running buses faster is a good idea. Right now… apparently it is not evident to Metro, for whatever reason.

      16. With regards to short distance vs. long-distance trips, I am concerned that any fare restructure that encourages more short distance trips is going to lead to significantly increased dwell time at bus stops, as more short-distance trips means every bus stop is going to have a lot more ons and offs, which means buses that move even slower than they currently do.

        Once the distance gets sufficiently short, particularly, in dense neighborhoods like Capitol Hill, transit competes for short-distance trips more with walking than with driving anyway – so lower fare to encourage short trips end up keeping people off the sidewalks, rather than cars off the road.

      17. asdf: That’s why you have a base fare (I started at $1). The point of that fare is to account for the cost of slowing the bus down for entry and exit, and to discourage people from riding the bus for extremely short trips. Ideally, Metro would pick a base fare (and a level of included mileage) that would avoid the situation you describe.

        Also, as a programmer, something I’m fond of noting is that constant-factor optimizations can have a huge benefit. If you have a bus that stops every 1/8 mile, and you delete half of the stops, you’ve just saved 50% of the dwell time that you’re ever going to save. Even if you deleted every single remaining stop on the bus, the time savings would be no greater than the savings from the first stop consolidation. So if you have a popular bus with reasonable stop spacing, then in practice, I doubt that the fare structure is going to lead to significantly more stops than what we currently have.

      18. Aleks,

        I appreciate your proposal as an interesting thought experiment and a stoker of discussion about the value of transit service at various speeds and distances. Nevertheless, I’m experiencing an almost visceral aversion to a proposal that…

        1) would impose upon any intra-city trips a penalty greater than the already highest-in-the-nation bus fare for slow, terrible service those passengers suffer today. ($3.50 for Ballard to West Seattle is simply too fucking much.)

        2) would, by design, privilege extremely short, one-seat city-center journeys at the expense of promoting faster trips across the urbanized area or the creation of a comprehensive urban network with a significantly improved transfer experience.

        3) would actively discourage residents from purchasing unlimited passes and inclining to use transit beyond their single most common journey, especially if that most-common journey is within the downtown-Queen Anne-Capitol Hill discount radius.

        The fundamental difference between Seattle and other distance-priced networks, from the Shanghai Metro to the Paris RER… is that those systems fundamentally work!

        At short distances, at medium distances, at long distances, your templates for this proposal tend to be advanced, comprehensive multi-modal systems that provide a high level of service across their respective urbanized and urban-adjacent areas. There is no such thing as a six-mile trip that can only be taken using a disproportionately arduous suck bus. Distance is distance is distance in these places, and so a sliding scale that acknowledges the roughly proportional relationship between distance traveled and resources expended can pass the twin tests of logic and fairness.

        (And anyway, the entire city proper of Paris remains within reach of the base Metro/RER fare, while Shanghai’s base fare will easily get you more than 5 miles before it begins to rise.)

        As ASDF points out above, your proposal sets the most-discounted fare at such a short distance that it encourages bus usage for trips that most non-passholders (and many passholders) might currently choose to walk. This is not a feature: since most of these trips involve the last mile into and out of downtown, you encourage additional crowding precisely where buses are already at their peak loads. With more passengers present than at any other point in the route, the cost of unnecessary boardings in wasted through-passenger-minutes adds up very quickly.

        Those delayed passengers, of course, are the ones who you are now forcing to pay extra for their longer, non-privileged in-city trips.

        ——

        You and I have discussed differential fares before, so I’m sure you understand I don’t favor (and am, indeed, offended by the off-peak existence of) a flat-fare across Metro’s massive service area. We’ve posited the Boston model as successful and easily replicable before: a flat fare for fundamentally local, urban-shaped service; a premium fare for “inner” commuter-specific express services; an even higher fare for “outer” commuter services. This acknowledges that specialized faster/further services (and their generally well-heeled users) should pay for the resources they gobble, while also encouraging a high rate of pass adoption and all-purpose usage within the core all-day network.

        ——

        Again, Metro charges the highest local bus fare in the nation, while offering a perpetually crappy service that is so structurally broken that the fare must continue to rise far faster than the quality. You attempt to triage this by privileging the very shortest trips, but your plan can’t help but raise the fare for many other basic trips out of the realm of the acceptable. By depressing the demand for cross-city trips, you reinforce the notion that transit is only ideal for one’s “primary” commute, while the personal auto is for all other trips, thereby rendering the structural revenue problem worse and ensuring that the cycle of fare hikes continues. It won’t be long before your Queen Anne base fare is back to $2.25 and the Greenwood trip is now up to $5. And our system will still suck too much to justify that.

        What I wish to see is a politically-achieved, permanent, non-volatile revenue stream, paired with a real and binding promise to streamline the urban network to make it work better for urban trips near and far. Then I’d like to see the fares frozen, unable to rise again until a slate of origin-destination pairs has dropped below ambitious trip-time thresholds (including waits and transfer penalties), until off-peak mode share has risen significantly, and until a certain percentage of city residents are passholders (a.k.a. stakeholders).

        The sign of transit system health is not how cheaply you can go to somewhere very close, and only slightly faster than walking, but how many places you get in a manner that would be worth your full $2.25. That is called “mobility”. That is transit freedom.

      19. I think we pretty much understand each other at this point, and I’m totally fine with not convincing you. :) There are just a few more things I want to say.

        – I obviously agree with you that, in terms of network design and operations, Metro has a lot to learn from the other systems that we’re talking about.

        – I think you’re putting too much weight in the particular numbers that I chose. Between Shanghai, Paris, Singapore, Hong Kong, and thousands of medium-distance transit services in North America — including our own Link and Sounder — it should be clear that there are tons of ways to calibrate distance-based fares. For example, in Singapore, the minimum fare covers about 2 miles, while the maximum fare covers all trips longer than 25 miles — and yet the maximum fare is only about 2.7x the minimum fare. (For cash users, all fares are higher, but the maximum fare is only 2x the minimum.) You could easily design a system where the longest possible in-city route (think Shoreline to Rainier Beach) was no more expensive than it is today.

        Coming up with the optimal numbers for King County (or even a close approximation) would require much more careful work than I have the time or ability to do. But any such system, even one where the Shoreline-Rainier Beach trip costs $3, still accomplishes (to some degree) the two goals that I care about: it increases the attractiveness of transit for short (2-3 mile) trips, and it stops subsidizing ultra-long trips at the expense of the core frequent network.

      20. While a $1 base fare may be sufficient to compensate Metro for the added operational costs as a result of one more passenger getting on or off the bus, it is not sufficient to compensate the other passengers on the bus for the resultant service delay.

        For example, if the bus contains 30 passengers who each value their time at $15/hour, and the delay factor of one boarding+one exiting is 1/100th of an hour, the total monetary cost of the passenger delay is $(30*15/100) = $4.50, much, much higher than $1.

        If we want more short trips, we either need a system that is more scalable than buses (e.g. Link), or a fundamental switch from large buses carrying lots of people to more frequent smaller buses, each carrying fewer people. Unfortunately, while the latter is done all over the place in the developing world, labor costs make it too expensive to implement in the United States (at least until bus driving can be automated by computer).

        Also, d.p. is absolutely right about the death spiral that your proposal would create for longer trips, forcing fares for shorter trips right by up again to make up for the lost revenue. In the long run, even people with employer-provided passes would be affected as employers would be less willing to provide such passes with increased cost. Raise fares to the point where a company like Microsoft is no longer willing to provide passes, I can almost guarantee ridership on routes like the 545 would fall off a cliff.

    4. Distance is predictable and consistent. Travel time is not. Should rider A have to pay extra because rider B held up the bus fumbling for change and Asking for directions? Should a crosstown commuter have to pay a $10 fare because it took 2 hours to get from the CD to Ballard today?

      imperfect as it may be, distance is far fairer than time.

      1. Yet at the same time, there are problems with a distance based fare in our geographically constrained region when distance is defined by as the crow flies. What should the fare be for a trip from Kirkalnd Transit Center to Sand Point compared to Redmond to UW or Bellevue to downtown Seattle?

        How about Alki Point to Magnolia versus Admiral Junction to Beacon Hill? Or Eastlake to Freemont.

      2. The system would already know the locations of all stops. If some very generalized data on shapes of the large bodies of water was created along with some extra points for the bridges across these bodies, instead of a pure straight-line path, the fare could be calculated as the shortest path connecting any number of points that also does not cross the banned regions (e.g. the three sections of Lake Washington). This removes the pen-paper fare prediction method though.

        It’s a problem that doesn’t necessarily need to be solved though. The number of perceived unfair start-end pairs is probably small and the difference in fare would also likely be small.

      3. The “number of perceived unfair start-end pairs” is probably equivalent to the number of riders of the system. :^)

      4. As Jarrod says, you could calculate the shortest-path distance between the start and end points, given the street grid, intermediate bodies of water, etc. At that point, it’s mostly a trade-off of fairness versus simplicity. I don’t have a strong preference between the two — I think both approaches have merits and flaws, and both are largely better than what we have now.

      5. I once operated a small transportation business in Portland using Mapquest’s “Shortest driving distance” function as the basis for fares. Probably not directly useful for a transit agency, but it shows that there are options out there that take geographic obstacles into account.

  2. I love this. This creates the right incentives for riders and Metro as well. Long-distance trips with good ridership will continue and even improve in service. Short urban trips that already have high fare recovery will get add even more riders. On the rider side, there’s no longer a suburban rider subsidy – you have an incentive to live closer to your work.

    We already have a tap-on, tap-off system on Link, so it would create consistency between modes. Everyone would tap-on, tap-off always.

    The trick would be with cash riders, but we’d just make them pay for the full route. This would add yet another incentive to convert to ORCA.

    1. Alex’s proposal sounds like a sheep in wolf’ clothing.
      Just another money grab by Seattle, at the expense of most of the population of King Co. paying the bills.
      Nice Try!

      1. Why is it always us vs. them? Everyone’s paying the same in taxes, just fewer suburbanites take the bus. Drawing the lines a bit differently, one could say that suburban transit riders are the ones riding on the backs of the non-riding suburbanites and their taxes.

      2. It’s not even suburban versus urban riders. It’s long-distance versus short-distance riders. Most B Line riders would probably see a fare decrease under this proposal, for example. Likewise for riders on the 140/F.

  3. The correct solution is a payroll head tax on all businesses (or an income tax, but that would be even harder to pass) and a flat $0 fare for all.

    Any system with variable fares causes an undue hardship on low-income riders due to the tightness of their budgets and the unpredictibility of the fares.

    1. No. For the same reason even Cadillac health insurance plans have co-pays, transit must have a fare. “Free” introduces extreme distortions in the demand of a service.

      Besides, from a practical point of view, can you imagine the uproar if transit were free but all the bridges into and out of Seattle were tolled? Eyman would have an initiative drafted faster than you can say “class warfare”.

      1. All of the transport infrastructure should be covered by that sort of progressive tax. No fares, no tolls, no gas tax.

      2. I’m not a fan of flat car tabs, high fares, or sales tax, either. But we need the bus service to exist, and don’t have other funding sources available right now. I consider today’s topic about an ideal future, but for the time being, I hope you will vote in favor of keeping the bus service funded.

    2. The Utah Transit Authority study that I linked to found that *flat* fares cause undue hardship on low-income riders, because they tend to take a higher number of shorter trips than higher-income riders.

      I don’t agree that fares would be unpredictable. If you want to know how much your trip would cost, you just check the distance on Google Maps. More importantly, if you take a particular trip on a regular basis, you’ll probably remember how much that trip costs.

      I do see your point that budgets can be tight. For the poorest riders, I think that the low-income ORCA is the right approach. If we did implement distance-based fares, I think it could make sense to maintain a flat fare for low-income riders, or even a daily maximum amount of money spent.

      1. Many of the low-income would need to take a bus to the library to be able to look up the distance in google maps, so infrequent trips are not predictable for them.

        If some sort of variable fare were introduced, a flat low-income fare would be needed to assist those eligible. Good point.

      2. I believe it’s even possible to figure out distance with a paper map. Or at least that’s what I remember from the days before Internet.

      3. I would prefer a discount on a distance based fare. We still want the same incentives, we just want them adjusted for hardship.

      4. Interesting that you bring up the Utah Transit Authority, where they are doing pioneering work (as US transit systems go) on accepting debit and credit cards for fare payment. That brings up the question of how easy it will be for private debit/credit card vendors to make their smart cards work with distance-based fares.

  4. Try explaining to a newcomer how to pay his/her fare. There is no simple way to implement this. With Link/Sounder, it’s tap-on, tap-off. If you forget to tap off, you pay the highest fare. For this scenario, would that highest fare be $9 (Seattle to Black Diamond)? Is it reasonable to require someone traveling from Downtown Seattle to have at least $9 on their ORCA card in the event that they are going far or forget to tap off?

    Or maybe you would tell the driver how far you are going. Aside from problems associated with the good old fashioned honor system (where people might exercise dishonorable behavior), there is an additional mandatory interaction between every passenger and driver that slows down the bus more (translation: COSTS MORE).

    How about more zones? Seattle, South King, North King, East King? That’s the most realistic solution. It’s really not all that different from the current system. I think the system we have is the system we have for a reason (although it would probably be prudent to have a two-zone fare of $2.50 off peak, and maybe a peak youth fare as well).

    1. Perhaps you pay the maximum fare for that route, or for that class of buses? So, every inside-Seattle bus would set their box at the maximum fare that any intra-Seattle route charges, every cross-lake route would set their box for that max, and only the buses going out to Black Diamond would set their fairboxes at $9. Of course, this would add another layer of complexity.

    2. For ORCA, I think tap-on, tap-off is probably the best we can do.

      Note that most of our longest routes aren’t actually that long. For a bus like the 143 to Black Diamond, pretty much no one will board in Seattle unless they’re riding across the non-stop segment to Renton, so no one’s going to pay $9 for what should be a $1 trip.

      As an example, the 358 travels 12 miles from Aurora Village to downtown. Therefore, the maximum fare would be $3.75, which isn’t much higher than the $3 that you’d currently pay if you were traveling that far during rush hour.

      The problem with zones is that they create artificial areas of extremely low bus usage. It’s not much more expensive to ride the bus across the Fremont Bridge than to ride it across 145th St, but the latter trip costs more. Of course, the fare differential between zones is tiny, and so this isn’t a huge disincentive. But any zone system will either have these inefficiencies, or else it will be so complex that it’s basically a distance-based system in disguise.

      To me, zones seem like a solution for implementing distance-based fares in the absence of smart cards. Now that we can measure the precise distance between your start and end points, I don’t see any way that zones would be an improvement.

      1. +1 to all of this, especially the “absence of smart cards” issue. We shouldn’t automatically assume the old way of doing something is good when we have different tools now.

      2. Singapore uses a distance-based fare system that is outwardly (i.e. for cash payers) insanely complicated. In practice, however, it drives what I imagine is near-total acceptance of smart card use as with the tap-on/tap-off system you don’t have to calculate it yourself. (As I recall there is a fare card on the bus for that route if you need to make the calculations.) Even as a highly experienced transit user overseas, I didn’t even try to begin to deal with the cash payment/distance issue there but did what everyone else did–get a smart card and tap in at the front door, tap off at the rear. http://www.smrt.com.sg/Buses/Fares.aspx

        The vast majority of people will inherently know the fare for their typical trip. The fare card onboard will help if you are fare conscious and taking an unfamiliar journey; of course this is all available online as well.

        They don’t bother to round the numbers either except for cash, where you pay a surcharge…you can get some very odd fares if you pay by smart card! It seems as though they may be doing what Aleks suggested by actually setting a farebox return percentage goal and setting per-km fares to meet that.

      3. Oran Viriyincy: “While they are keeping the zones for now, they do hint at possibly moving to a distance-based fare structure in the future.”

        I live in Metro Vancouver and if TransLink wants to do that they would need approve from mayors, the Provincial government (I don’t about that) which would come from the Ministry of Transportation, and TransLink indepent commission

      4. Can edited please delete my last edited because it wasn’t allow me to do anything.
        Oran Viriyincy: “While they are keeping the zones for now, they do hint at possibly moving to a distance-based fare structure in the future.”

        I live in Metro Vancouver and if TransLink wants to do that they would need approve from Mayors, the Provincial government (I don’t about that) which would come from the Ministry of Transportation, and TransLink independent commission would need to agree to fare structure change. If one says no to a change in a fare structure change they will not be able to change it legally.

      5. I’m concerned that a tap-on-tap-off system with buses to implement distance-based pricing is going to lead to massive fare evasion. It’s just too easy to tap on when you board, then immediately afterward tap off, without actually getting off the bus.

        Link doesn’t have this problem for multiple reasons. First, the card readers are at the stations, not on board the train, which means if you get off the train at an intermediate stop to do a fake tap off, the train will probably leave without you. Second, Link has random fare inspectors and if they see you on the train after you’ve already tapped off, you’re going to get busted. If you’re going on take your chance on not getting caught, you may as well not bother tapping on in the first place.

        With buses, off-board fare payment is not practical at every stop because there just too many bus stops. Nor is it financially viable to hire fare inspectors for every single bus route.

      6. asdf: I understand what you’re saying. But, at the same time, this system is already used by Singapore, and will soon be used by Vancouver. Singapore’s transit agency is actually profitable, so it’s clearly not leading to mass fare evasion there. The Vancouver experience will probably be a much closer analogy to Seattle; if they succeed, it’s reasonable to assume that we will, too.

      1. That +1 was for Alexkven – zones work fine but there has to be a good map readily available showing the first and last stops in each zone on each route. A much more practical solution to better fare recovery.

  5. First, let me say that I am mostly interested in comments from those who preface their comments with “I strongly support [Metro’s] efforts, and will enthusiastically vote for the measure when it hits my ballot.” Sans that, the discussion just doesn’t feel like something in the real world.

    Second, I strongly support Metro’s efforts, and will enthusiastically vote for and campaign for the measure.

    As a consumer of transportation services, there are three basic things I am looking for in pricing:
    (1) that I can accurately and precisely predict how much I will pay for a trip *before* I take that trip;(2) that the method of payment collection does not get in the way of the speed of the trip; and (3) that I can afford said price.

    Distance-based fares fail criterion (1), unless I have a handy chart (such as is available for Link, Sounder, Amtrak, or an airline) showing me how much it costs to ride from my point of origin to my destination. When criterion (1) fails, then I don’t know the answer to criterion (3). If I am cash-strapped, I might then decide not to take the trip, and the fare system has just defeated my access to transportation. For reasons of failing criterion (1), I avoid taxis like the plague (and that, from my experience, they cost a few orders of magnitude more than a bus ride).

    A double-tap system makes the most sense when it is off-board, and fare disputes are handled off-board. I’d be curious to hear where an on-board double-tap system has been implemented, and how early-second-tapping is detected and punished.

    Metro has the rudimentary beginnings of a distance-based system with its 2 zones, but its zone-based and time-based system fails my criterion (1) when I am on the margin of peak hour (whenever that really is implemented in practice) and when I am crossing a zone boundary. Really, it just punishes riders for living near the Seattle City Limits, which just happens to be where there is a high concentration of low-income housing.

    Having a system where riders may end up paying less in practice than policy says they should have paid neither helps Metro’s bottom line, nor does it help poor riders who decide not to take a trip because they thought they couldn’t afford it, when in actuality they could have.

    As to the implementation specifics, the two-tap system for Link and Sounder is straightforward in its implementation because the individual readers are off-board and set to their specific location, and the amount charged is based on an internal chart of the location-to-location prices, rather than an actual distance computation for each transaction. An off-board double-tap system would also work for RapidRide if only every stop had an ORCA reader. Implementation of on-board double-tap would be much more complex, involving GPS, or perhaps frequent driver attention to reset the stops, as well as having a mechanism to shut off the readers when the fare inspectors board (which will have to be faster than the fare cheaters sitting/standing next to the reader).

    FInally, let me close with a Sam-level question, Aleks: Is “distance” defined according to how the crow flies, or according to the odometer?

    Thanks for your post, Aleks, and I hope to run across you on the campaign trail!

    1. Brent,

      Thanks for your thoughtful reply!

      I agree with all three of your listed criteria.

      I’m going to answer your last question first. I think fares should be based on distance as the crow flies, not as the bus travels. As I mentioned above, if I want to get from Capitol Hill to the International District, should I pay extra because Metro doesn’t run a bus down Broadway? The difference between GPS distance and odometer distance mostly comes down to routing choices, and since only Metro can change bus routing, Metro should be bearing that cost, not passengers.

      Given that, I would argue that distance-based fares *do* meet criterion #1. Time-based fares completely fail, of course, since you would never know how long you spent on the bus until you actually spent the time. (This is the same problem that taxis have, as you describe.) But with distance-based fares, you can ask Google Maps, and it will tell you the exact distance between two points. You could also imagine having an app, and a page on the Metro website, where you can type in your origin and destination and it will tell you the exact fare.

      It’s certainly true that there are some passengers who won’t have a mobile phone and thus won’t be able to predict their fare. But I don’t think the situation is any worse than the situation for air travel, which you seem to suggest is acceptable.

      Regarding how it would be implemented, Oran’s Singapore links are spot-on.

      1. As-the-crow-flies distance based calculations look good for one-way trips, but they price round-trips way too cheaply.

        For instance, if you’re making a 5-mile-each-way run to a store, your system would result in a round-trip fare of the minimum $1, since you end at the same point you started with, leading to an as-the-crow-flies distance of 0.

        And if you’re proposing to solve that problem by special-casing trips that end at the exact same stop they start, it won’t work. That measure could easily be defeated by simply getting off the bus one stop earlier on the way home and walking a few blocks.

        Somehow, system would have to distinguish between out-of-direction travel that constitutes a legitimate round trip with out-of-direction travel that is merely a case of an undesirable transfer due to a bus system that sucks. Unfortunately, making such a distinction without reading the passenger’s mind is virtually impossible.

        For example, let’s consider two people, Alice and Bob, who are each busing one-way from Fremont to Lake City via downtown. Alice lives in Lake City and is returning home from a visit to a friend in Fremont. She got to Fremont because her friend drove out to Lake City to pick her up because the bus service between Lake City and Fremont is so awful. But, on the way back, her friend was busy and didn’t have time to do that drive again, so Alice has to go downtown and catch the 522 (waiting a half-hour downtown between buses) because that’s what the bus network forces her to do.

        Now, let’s consider Bob. Bob also lives in Lake City, and is also coming down to Fremont to visit a friend. Bob owns a car and would normally be driving the round trip because of the poor bus service between Fremont and Lake City. But, in this case, Bob has another friend from out of town staying at a downtown hotel and wants to combine the visit with the first friend with a visit to the second friend. So, to avoid the hassle of parking downtown, Bob drives Car2Go from Lake City to Fremont, then rides the 28 from Fremont to downtown to have a quick dinner with his other friend, then hops on the 522 back home to Lake City.

        Both Alice and Bob have made virtually the same trip, spending a similar amount of time downtown between buses. Yet, for Alice, the trip is an awful transfer while, for Bob, downtown is merely an extra stop on a multi-leg journey, and the two legs really are, conceptually, two trips. There is no way the computer could charge the “fair fare” to both Alice and Bob using your scheme.

      2. Aleks,

        Google Maps will by no means always tell you “the exact distance between two points”, if by “exact distance” you mean “as the crow flies”. It will tell you the exact distance as the blue line goes.

        Now if that blue line is straight, sure, it’s the crow flies distance. But how often is that true? Any turns immediately introduce extra distance.

        It would be great if GM added an option for geographical distance, but I at least have never figured out how to turn it on.

        @asdf,

        The hypothetical difference between Bob having a convivial dinner downtown and Alice waiting at some cold blustery bus stop is a bit dramatic. It’s likely that Alice would find something to do herself.

        Anyway, since when should government charge people different prices for the same activity based on the “motivation” for that activity? Talk about a slippery slope!

    2. I strongly support Metro’s efforts, and will enthusiastically vote for and campaign for the measure. :)

    3. Agreed regarding (1). There’s a reason zone fares are so popular as a system — they are mostly distance-based, but they’re easily comprehensible.

    1. or with another ‘elite class’ of transit rider. Those that work for larger CTR employers, and get a Passport – good for anything to anywhere at anytime, and all highly subsidized by both Metro and a few employers willing to participate and enjoy some additional tax breaks.

      1. The reason why Metro is selling these passes at a fraction of the face value is that the business is required to offer the passes to every employee, for free, in order to receive the discount. It is absolutely expected, especially for suburban companies, that many employees will take the bus pass because its free and only use it to go downtown once or twice a year. If you look at the total cost paid per actual use, the discount is not nearly as much as it seems.

        That is also why suburban employers get the biggest discount – because the number of actual trips taken per discounted pass is usually much less.

      2. the business is required to offer the passes to every employee, for free, in order to receive the discount.

        This is not true. From Metro’s ORCA Passport FAQ:

        Q: Can our company share the cost of the Passport with our employees?
        A: Yes. The employer agrees to pay for a Passport program based on the full number of benefits-eligible employees at the company. The company can then choose to have employees who receive the pass co-pay up to 50% of the annual cost of an individual pass.

    2. I don’t think monthly passes would need to be much different than today. You can buy passes in all denominations, and the cost is 36 times the per-ride value of the pass. If you have a few different trips you take regularly, it doesn’t make sense to buy a pass for the more expensive trip if you plan to take that trip more than 18 times (round trip) in the month.

      For example, suppose you ride Metro every day to get to work. At least half the time you travel outside of peak hours, but sometimes (say 20 one-ways each month) you travel during the peak time. You could buy the $2.50 pass to cover the peak trips, but that would cost you $90. It’s actually cheaper to buy the $2.25 pass for $81 and pay an extra quarter out of your e-purse ($5/month) each time you ride during peak hours. The system is already set up to handle this.

      I think a more interesting question is what would happen with transfers under a distance-based system. Right now, you’re charged for the most expensive individual leg of your journey. If you take a $2.25 Metro ride to downtown and then take Link to the airport for $2.75, all you end up paying is $2.75. That policy could have some interesting consequences under a distance-based Metro fare collection system. Suppose I’m in the middle of a long bus trip and I can see (from OneBusAway or other means) that the next bus in the same route is just five minutes behind me. Can I get off the bus, wait five minutes for the next one, and end up paying half price because I only rode each vehicle for half the overall distance?

      1. As I mentioned in another comment, I would like to see trip cost based on end-to-end distance as the crow flies. If you’re trying to get between two points, and Metro forces you to go out of your way (e.g. downtown) to get there, it should be Metro who bears that added cost, not you. (Airlines would never charge you extra for an inconvenient layover; they give you a discount, instead.)

        So if you connect to two different services, Metro will simply ignore the taps from when you got off the first vehicle and when you got on the second one. Your resulting cost could end up being as expensive as if you rode one bus for the whole trip. Or, if you’re heading in a different direction, it could end up being cheaper (since you’ll only pay for the hypotenuse). But you can’t game the system by exiting and reboarding the same bus route.

      2. Would charging fares based on bus odometer distance be an incentive to get people to advocate for more direct, less circuitous routes though?

      3. Would charging fares based on bus odometer distance be an incentive to get people to advocate for more direct, less circuitous routes though?

        That’s too many levels of indirection for me. Metro has direct control over its route network; passengers do not. The incentive for Metro is much more direct and much more powerful. And if it takes Metro a year to put together a reorganization, propose it, and get it approved, it’s unreasonable to ask passengers to suffer in the meantime.

        Again, I’ve never heard of an airline that charges less for non-stop routes than for routes with a layover…

  6. “No other transportation service works like this. Taxi fares are based on distance and time. Toll roads cost more the further you drive. Airlines set fares based on costs and demand. Long-distance trains and buses, commuter rail… the list goes on. Only local transit networks have flat fares.”

    I can think of one more transportation network with flat fares: a private automobile. My car tabs cost the same whether I leave my car parked at home for a year or drive it in a continuous repeating z-shaped course from coast to coast and border to border. And thanks to the defense expenditure called the interstate highway system, without paying a dime in tolls.

    What my license plate tabs and my monthly transit passes- which used to be available for a year- buy me is permission to travel as I please. The sense of freedom, l especially from one more small nuisance in my life, is literally priceless.

    The rear-door card readers should have gone aboard every vehicle in transit’s every fleet years ago. Every single door on every San Francisco MUNI bus and light rail car has one. Fact that King County Metro failed to follow suit last time it had the chance is an excellent argument for finding another agency to run the system. Would be good if voters made their “yes” vote conditional on the County’s ending every policy that makes transit more annoying and slower.

    But my main quarrel with your main worry about social inequity is that across the income spectrum, most people now live regional and national lives- and have been willingly doing so beginning in about 1915, when streetcar ridership winged-over into its fatal nosedive nationwide as working people could finally buy cars.

    Good metric for the revolution we need now would be a hundred years of employment for our whole population that paid enough to buy our automobiles out of our earnings, debt free. My guess is that the level of education to make that possible will ensure that this time, people will freely choose to spend the money on public transit instead.

    Mark

    1. “I can think of one more transportation network with flat fares: a private automobile. My car tabs cost the same whether I leave my car parked at home for a year or drive it in a continuous repeating z-shaped course from coast to coast and border to border.”

      Yes, but then you also pay gas tax based on how much you drive.

      1. What gas tax? Electric! ;-) OK, yes, there’s a variable fuel cost either way, and there are taxes on electricity, too.

        Anyway, the same “flat fee” character is true for bicycles, only there the fuel cost is completely invisible to the cyclist (unless he or she loses weight). So it looks like a flat fare to a very extreme degree.

        Interesting to think about.

    2. I definitely hear what you’re saying.

      The converse is that the very same interstate highway system is arguably one of the reasons that Americans lead increasingly disconnected lives; that our cities spent the better part of a decade crumbling and depopulating; that a $15,000 device (a private automobile) became a necessary purchase for even the poorest households.

      We’ve all spent the past 50 years thinking that distance doesn’t matter. But why are we seeing an urban renaissance, if not for the fact that distance *does* matter? Closer means that you can walk to the grocery store, and get to work without a car, and have fun in your neighborhood without ever being tempted to drink and drive. Closer means that you can always find something to do, or something to watch, or someone to help you, wherever you are in the city.

      I think a bus system that subsidizes long trips at the expense of short ones is fundamentally anti-urban, and that’s a strange characteristic for a bus system to have. And yet, that’s what we’ve got now.

      One last thing. As RedmondRider says, drivers *do* pay more when they drive further. The IRS estimates the variable costs of driving at 24 cents per mile. It’s true that many drivers don’t think about costs that aren’t gas, but they still exist. Frankly, riding the bus for almost the same per-mile cost as driving feels like a pretty good deal — you’re basically paying someone a penny per mile to drive you anywhere you want to go! :)

      1. “We’ve all spent the past 50 years thinking that distance doesn’t matter. But why are we seeing an urban renaissance, if not for the fact that distance *does* matter?….”

        “I think a bus system that subsidizes long trips at the expense of short ones is fundamentally anti-urban, and that’s a strange characteristic for a bus system to have. And yet, that’s what we’ve got now.”

        This is a really good point. I could call it a manifesto, perhaps. I may quote you on this in the future…

  7. What about the elasticity of demand? If you raise the price of something, you’d expect people to buy less of it. I don’t see anything in your analysis taking this into account. On the face of it, I’d expect long commuter routes to have a higher elasticity, since more of the riders have the choice to drive instead. These are the same routes that you’re proposing to charge a lot more on. So if ridership on these routes fell significantly, that’d have a big impact on your projections. Would you really be able to meet the revenue shortfall then?

    1. The thing is, those same commuter routes are also some of the most expensive services that Metro runs (on a per-trip basis). So if ridership goes down on those routes, then Metro just cancels trips, and that saves money.

      Conversely, for the same reason, you would expect short-trip ridership to increase. If a lot of riders start using Metro for trips for which they formerly would have walked or biked or drove, that could end up raising a lot of money, without hugely increasing costs (since the buses are already there and generally not overcrowded).

      I don’t think we can really anticipate the full effect on ridership without doing a much more comprehensive study than my cursory evaluation. If Metro wanted to seriously investigate a distance-based pricing system, they would conduct a number of surveys, and figure out the appropriate level for fares. As an example, Sound Transit recently did this for Tacoma Link, and WSDOT did it for 520 and the AWV tunnel.

      1. You’d want to allow more density to account for the increased demand this would cause for living close to work. :)

      2. Aleks,

        If you end up canceling a bunch of suburban expresses, you end up with an increasingly hostile-to-transit voting public. Even folks who live in the service area but don’t use the suburban expresses like that fact that they’re available for days when the car has to go to the garage and for their kids to use.

        So it’s not a good idea airily to waive away the effects of stopping that $9 trip to Black Diamond.

  8. I have a question about how you define distance. Are we talking geodesic, bee line distance? Or would it be based on how many miles the bus has traveled? For example, if I take the route 75 from the U Dist. to Northgate, that’s a distance on the map of about 2 miles. But the bus has probably traveled 6 miles by the odometer getting there. Would I be charged for 2 or 6 miles?

    Thank you.

    1. That’s amazing! Brent asked the same question, calling it a Sam-level question, and now here you are asking it yourself! ;)

      I think it should be based on bee-line distance. An airline would never charge you more because they make you deal with an inconvenient layover; if anything, 1-stop fares are generally cheaper than non-stops. Similarly, passengers shouldn’t have to pay more because Metro’s routing decisions cause their trip to be longer than it has to be.

    2. Aleks stated in an above comment (shortly after you posted this) that he would prefer straight line distances.

    3. I believe Sam reads my comments before he browses the NYT crossword puzzle, because he finds my comments more compelling.

  9. That is what I like about Hong Kong’s system.

    Our trunk routes, or local routes as you call it, that stop more frequently charge a lower whole fare than limited-stop and express routes. The sense is that local routes are for people who ride short distances (25 minutes on local routes. Express route riders enjoy a faster service, and they generally ride for longer distances. These should be paying more.

    Let me give you an example.
    The Hong Kong Island route 2 and route 720 both run between the east side of the island and the CBD. However, route 2 runs on mostly local roads, while 720 runs on the highway. Route 2’s target customer is not an east-west rider, but rather a short-trip rider. 720 targets east-west riders primarily. It takes 55 minutes (avg. sp. 14kph) for a route 2 bus to finish a trip, while it is just 34 minutes (avg spd 20kph) on average for 720. The whole trip fare for route 2 is HKD$4.1 (US$0.53) versus 720’s HKD$6.5 (US$0.84). It looks like route 2 is not profitable. However, as long as there are more than ~2 bus full of boardings per trip, route 2 can potentially make more money.

    The fares of bus routes in Hong Kong are set according to a list called “Scale of Fares”.
    http://www.gld.gov.hk/egazette/pdf/20131709/egn201317091131.pdf

    There are different route types. Longer routes charge more, but less in terms of fare per km. Express routes also charge more. After the express section, the express routes will charge the same price as local routes operating on the same routing.

    I believe that is quite a fair system. It also eliminates the need for a monthly pass, where people can ride for unlimited boardings, wasting time and money for the TA. We don’t have monthly passes in Hong Kong. Monthly passes are actually quite unfair to non-pass riders, in my opinion.
    Why should someone pay less than you do while he/she can travel for a longer journey than you do?

    1. Some correction there:
      Local routes are ideally for passengers travelling for 25 minutes if it were a local service.

      In summary, what I think is fair is to have local routes charge a slightly higher $/km trip fare, but the total fare remains slightly less than the equivalent express routes. Just like the example I used above.

      By the way, the fare box recovery rates of North American TAs are quite scary…

    2. A colleague of mine, and an occasional poster here, once suggested to me about making express routes here 1$ more. You are paying for a faster trip (fewer stops). I wonder how much money metro would raise from that idea

      1. I think lowering the local fare to $1.75, keeping the limited fare at $2.25, and raising the express fare to $2.75 (for shorter routes) and $3.3 (for longer routes such as 952) would be a good option.

      2. CERTAIN express routes should have premium fares, namely the long-distance peak expresses. Not the regional trunk routes or short-distance expresses.

        Metro’s problem is it’s harder to institute that after the fact than at the beginning. It originated as a rural county-based system, with peak-expresses to all those areas, and a two-zone fare structure. The easiest solution now would be to raise two-zone fares, but that would hit people who live near the zone boundaries and may even be on local buses (5, 106. 120, 131, 132, 150, 358). So Metro really should single out the longest peak-expresses for premium fares, but that’s politically difficult.

        Plus it runs into complications where two routes would have vastly different fares for going to the same place. This already occurs where Metro routes shadow ST Express routes, but they’re essentially overflow because ST couldn’t cope with the volume if Metro cut those. (E.g., the 218 is full.) I can see those as part of the trunk network, so regular fare. What burns me up is peak-expresses going directly from single-family neighborhoods to downtown without a premium fare.

    3. Because that person is a routine rider, who has “bought into” the system on a long-term basis*, who has entrusted the system to a greater degree with his or her personal mobility, who has expressed a willingness to use the system holistically and to step away from any automobile-oriented cultural default.

      In my experience, at least in Western counties, the most robust and ubiquitous transit systems tend to see 70%-85% of all patrons riding with some form of time-based unlimited pass. By contrast, systems with high fares and no discount for routine users suffer from grossly under-capacity off-peak ridership (see: BART). Why anyone would eschew the former for the latter is beyond me.

      *(This is somewhat undercut in cities like Seattle, where a large number of passholders have received them as employment bribes, and who are therefore divorced from feeling any stake in a system whose quality justifies the face value of the pass, or that succeeds at more than expressing them at 8am to their downtown cubicles).

      1. Hong Kong manages to make bus trips cost less (avg. ~US$1) than Seattle’s, even with bus pass taken into account (assuming fair, twice a day usage). We don’t really have out-of-reach fares like you get in Seattle.

      2. That’s a fairly irrelevant comparison, given differences in labor costs and the very large “captive audience” in a city with negligible rates of car ownership.

        However, Boston manages to charge only $70 a month (or a $1.70 subway fare) for a system with one of the highest percentages of monthly passholders in the U.S. That should better demonstrate the economies of scale that become feasible when a significant percentage of the urbanized population elects to “buy in”.

      3. Sorry, I didn’t word it correctly up there. A lot of Hong Kong’s bus trips cost less than the ~US$1 average. That is because we can pay much less than that on shorter trips. However, in Seattle, even if you have a bus pass, if all you need is a 10 minute ride, the fare is ridiculously high and off-putting. Now, if you consider a 358 or 311 passenger, who travels on almost the entire length of the journey, yet paying just slightly more than you do, it is very unfair.

        I think one important aspect is how much the passenger feels his/her trip is worth. If you have a very long route that costs the same to right as a fairly short route, I don’t think the passengers on the shorter route would like it. In addition, what kind of equipment you use is another decisive factor. You need to make sure that the ride environment makes the passenger feel that the trip is worth the price they pay. Poorly maintained buses would not be good.

        By making local routes cheaper than they are now, it attracts (potentially new) short-haul riders without bus passes to use the bus. If new riders try the system and become satisfied about it, you can increase the long-term ridership.

        A higher fare for express routes can be justified with a better spec-ed service, such as wider high-back seats, or even double-deckers. Probably try restricting a monthly pass fare class to a set of similar routes. This will make people pay a reasonable amount for the service they use, while still paying less than if they were to pay for the ride each time.

        In the end, you need to make sure that the passengers who are supposed to pay more for their trip do pay more, and vice versa.

        Another way to boost transit usage is to make parking in downtown excessively expensive, forcing people to use P&R services instead. Or even one step further, introducing a $3 daily downtown parking tax will give transit quite a good sum of money. Driving to downtown alone for work is quite a selfish act.

      4. You worded it quite clearly. You just happen to be missing the point.

        I’m fully in favor (as Aleks knows) of surcharges for express buses and premium commuter services, and for trips that unquestionably traverse great distances for purposes distinctly different than urban transportation (and generally requiring much higher per-passenger subsidies).

        But what you forget when you talk about “if all you need is a 10 minute ride”, is that this is not Hong Kong or Shanghai or London. Urban transit in Seattle is so fundamentally broken that “a 10 minute ride” is often than one mile, i.e. well within the walking threshold of anyone who doesn’t use transit frequently enough or widely enough to warrant a monthly pass for the whole urban zone.

        Until urban transit here ceases to be that broken, I take umbrage in my gut and in my soul to a proposal that could wind up charging urban riders more than they pay now to spend enormous amounts of time crawling 8 or 9 miles on those same slow-ass, fundamentally broken services.

      5. When I wrote 10 minutes, I was thinking more like the distance from Lynnwood 36th Ave W to Lynnwood TC or Ash Way. You can’t really walk between these places. I have tried personally, and I gave up.

        I think you have also missed my point. There shouldn’t be an unlimited bus pass that can allow someone to take unlimited rides on the entire system. A slightly cheaper bus pass that works only on a set of similar routes (similar endpoints), say 70, 71, 72, 74, 75, would be more financially viable for KCM, without making the passengers pay more.

        In your last paragraph, you mention passengers need more time than they should to travel 8 or 9 miles. That is crap route design. However, it also indicates that the route is a slow, local route, which I propose lowering the trip fare depending on the total distance. Again, take a look at these:

        http://www.gld.gov.hk/egazette/pdf/20081222/egn200812223586.pdf
        http://www.gld.gov.hk/egazette/pdf/20081222/egn200812223584.pdf

        This fare system is fair in the sense that it takes into account the type of the route, the distance of the route, and the service level of the route. As long as you offer the same price for similar routes, people will not be confused.

    4. “It also eliminates the need for a monthly pass, where people can ride for unlimited boardings, wasting time and money for the TA…. Monthly passes are actually quite unfair to non-pass riders, in my opinion. Why should someone pay less than you do while he/she can travel for a longer journey than you do?”

      For the same reason that newspapers have a subscription discount, wholesale grocers charge less for 10-pound and 50-pound bags, and the U-Pass has created a lot more transit in the U-District than anything else has. Passes generate a stable and predictabe revenue stream, and eliminate cash transactions. The agency has less expenses and more predictible budgeting, and rebates the savings on to users. That’s why Metro and most agencies encourage people to buy passes. (They could just eliminate the passes if they didn’t want people to use them.)

      People have an upper limit on how many trips they’ll take, even with an unlimited pass, because they have other things to do. (Except homeless people, but there the problem is a lack of day shelters, which would be more interesting than a bus seat.) It’s up to the transit agency to predict what people’s collective limit is, and to price the pass accordingly and have a sufficient number of buses to meet demand.

      What’s missing in Metro is a weekly pass, for those who can’t use a whole month at a time (especially if it can only begin on the 1st of the month).

      Unlimited passes give a psychological reassurance that goes beyond the physical bus trips. I buy a monthly pass (at full cost) ever month even when I’ll be away for a week or two or commuting by bike. That way I don’t have to agonize over “Is this trip worth $2 to me?”, or worse, get on a bus without cash forgetting I skipped this month’s pass. Plus it’s a predictable monthly expense which makes accounting simpler.

      1. @ Mike Orr – “Unlimited passes give a psychological reassurance that goes beyond the physical bus trips.”

        Precisely. I will always get the unlimited daily pass whilst traveling if that is an option, and I find myself using transit more than I might have otherwise (same goes locally). Not that the single fares can’t be mustered up; rather the feeling that I now have a “ride free” card and every bus or train is a potential ride somewhere. It doesn’t matter how far I might choose to walk if I feel that without issue I can just hop on transit back whenever I see it, if I so desire.

        It was always a lot more fun to have the day pass at an amusement park than to have to buy tickets for each ride!

      2. ]I support the proposal, but that’s easy for me: I have no car to worry about the price of tabs on.] Unlimited passes increase ridership, and they increase the chances of people stopping off somewhere: it suddenly makes more sense to go to that interesting-looking shop, or a side trip to the library. I have trips now where I think “hmm, I’m close to using up my two hours, should I take this side trip?” because it’ll cost me another $2.25 or $2.50. That probably helps support local business.

        I remember when New York city added free bus-to-subway and subway-to-bus transfers, they had an increase of something like 40% in bus rides taken in one year. A lot of that was people deciding to spare their feet/legs at the end of a long day, rather than walk the last 1/2 mile or mile home. (You don’t take most of those crosstown buses to save time, but they were a blessing when I was recovering from a knee injury.) I’m all in favor of walking as exercise, but let’s make it a choice, not something working people have to do even on cold, damp days.

  10. Problem with the back of the envelope math–assumes no lost ridership due to fare increases. This creates a number of potential problems beyond not hitting your revenue target. It could lead to loss of service, due to low ridership, and a general loss of political support outside of the core, dense areas. Long term, we need a broader base of support for transit than that, even if it means ‘bribing’ the outlying areas with cheap long-distance rides.

    1. You’re also assuming no gains in ridership due to fare decreases. This proposal would lower fares for many trips that are already very popular, as well as for many trips that *should* be more popular (like short hops in the suburbs).

      I agree with you that my proposal is a long, long way from a comprehensive plan for implementing distance-based fares. You’d have to do a real, formal study, and it would have to take into account the effect on demand. But I think it’s far from a foregone conclusion that it would automatically lower total ridership compared to a flat fare that raised the same amount of revenue — especially when you consider that a higher system farebox recovery ratio means that you can run more service for the same amount of tax dollars.

      Regarding outlying areas, I think this proposal would increase ridership on short suburban trips — for example, Redmond TC to Overlake TC, or Bellevue TC to Factoria. I think that suburban express riders have a greater capacity to absorb price increases than you might expect, because the alternative is paying for downtown parking; as long as their commute is cheaper than that, they’re better off (financially) taking the bus. But again, we don’t really know without doing the full study.

      1. And on the downtown parking point, the number of parking spots downtown is more or less fixed and completely consumed already. If a significant number of commuters even tried to switch to driving that would raise prices until they (or someone else) changed their minds.

        There are other negative effects that could happen (increase incentive for businesses to move out of downtown, or developers building new parking structures), but prices would have to rise quite a bit for these to be large effects.

      2. Whenever I’ve gone by Pacific Place, it’s had open spaces – but I don’t go by it too often.

      3. I was certainly making a jump based on logic, but I used to walk two miles partially to avoid a $14/day parking garage. Surely there isn’t anywhere downtown you can’t just drop a few dollars and fill your lot. Right? But looking at those charts, it seems like I’m wrong.

        But wait, I was talking about paid lots. I assume those charts have a lot of residential lots (empty all day) and office building lots (where it may not make fiscal sense to open them up to the public).

  11. If we are going to seriously try distance-based fares here, the E Line would be the perfect place to run a pilot project. The base fare could be the 1-zone fare, with the max fare being the 2-zone fare. (Any lower fare would be a hit on Metro’s budget, and therefore be politically DOA.) RRFP, youth, and low-income riders could stick with their flat fares.

    I would like to see such a pilot project, even if the only productive outcome is installing ORCA readers at all the E-Line stops.

    1. I think that’s a great idea! It would give Metro some experience with tap-on, tap-off for buses, but with relatively low risk. :)

    2. Anything would be better than the fare-inspection trap they’re planning for RR E. Although I’d prefer a flat-rate fare on it.

    3. Love the idea of distance based fares on the 358 (I refuse to call it Rapid Ride, because there is nothing rapid about it). It would actually have me ride the bus to work for a change. Right now, it costs $6/work day to get to work via the 358. I live 2.5 miles from my office, yet it is far cheaper for me to drive to work than commute via the 358. To add insult to injury, I end up paying two-zones even though my trip remains within Shoreline.

      1. You can tell the driver “One zone” and they’ll change the reader… if they can figure out how. I usually have the opposite problem. I tell the driver “two zones”, they press a few buttons, can’t get it to work, say they do it so rarely they don’t remember how, and tell me to just tap the one-zone fare.

  12. Metro already charges different fares base on distance (zone) travelled in addition to the time of day, rider’s age, and soon income, too. Is the goal here to make the fare system as complicated as possible?

    1. You know when you board the bus whether you have a low-income ORCA or an RRFP. So that doesn’t complicate figuring out the fare at all. Figuring out the flat youth fare is accomplished the first time you board the bus, and find out whether you need a youth ORCA in order to get the youth fare. (I’m embarrased to say I don’t know whether a simple ID gets drivers to accept the youth fare.)

      I wholeheartedly agree that zones and time-based fares are nuisance complications, especially when few have memorized the zone lines or know the rules about how the time-based fares are charged. Metro has data showing that poor riders are a larger share of off-peak riders than of peak riders, so they consider the “off-peak discount” to be a sideways way of addressing social equity.

    2. Short answer is YES. All the transit agencies have had 20 years to perfect fare integration and make riding transit seamlessly simple.
      Anyone want to guess how it’s going so far?

  13. One last point on social equity: If the goal is to make the bus system cheaper for riders who have trouble affording it, Metro’s proposed low-income fare program is much stronger sauce than UTA’s dabbling with distance-based fares.

    Don’t get me wrong. I think distance-based fares can achieve some important goals. But for social equity, there is no substitute for a discount program for those of low income. And as several have pointed out here, distance-based pricing lookup is much less accessible to those without pricey phones (which would include me). Moreover, UTA, for all its pioneering work, still hasn’t solved the problem of “unbanked” riders, since it has been reluctant to roll out a transit smart card for the general public.

    1. I absolutely agree with you that we need an immediate and direct solution for low-income riders.

      I do think this proposal would have equity benefits for middle-class riders — folks who don’t qualify for low-income discounts, but who aren’t Scrooge McDuck, either. The 545 is full of folks going to Microsoft; the 3/4 are full of folks going to Harborview.

      I also think this proposal would make it possible to provide more bus service for at any given level of tax revenue. I think that’s the best equity benefit of them all. :)

  14. Given that a Sounder ride costs more from Seattle to Tacoma than from Seattle to Sumner, it seems clear ST is using odometer-based distance measurement. If they switched to beeline measurement, and reset the fares in a revenue-neutral manner, it might incentivize more riders to opt to fill up the seats on the train instead of riding an express bus (with higher marginal costs).

    Or perhaps ST *wants* riders to choose the express bus over the train.

  15. What was Farber’s dataset when he made his findings? Because if it was just within the Salt Lake City region or Utah generally, I’m not sure it would apply to Seattle where land use patterns are quite different and gentrification is so prevalent. Especially if you exclude peak only commuter routes (which arguably should have a separate fare system anyway), I would not be surprised at all if trip distance was negatively correlated with income/wealth in King County. Moreover, providing low income passes and fares is a much more direct way to address the needs of low income people.

    In addition, as other commenters have noted, there would be a variety of implementation problems with a distance based system. Not that a distance based system would be totally out of the question, assuming it was measured by as the crow flies distance, but I’d want to see a lot more data to support the idea. In particular, it would be useful to consider thoroughly how it would effect land use, as incentivizing density and living near one’s job would be a good thing.

    1. I have to agree that Alex may have a point on the negative correlation, given how Seattle is full of millionaires in high-priced single-family homes, and lots of people have been pushed — against their will — to live in suburgatory. Most US cities I’ve seen are the reverse: “white flight”. While distance-based fares in these typical American cities may help achieve a small amount of fare progressiveness, I tend to believe the effect here in $eattle may be the opposite.

      I prefer flat fares for low-income riders over punishing poor riders for living where they could afford to live.

      1. I’m not an expert on Salt Lake City, but Seattle has one of the highest gentrification rates of large cities in the country and that shows in sky high housing prices within the city. Seattle has also has been relatively well off compared to its suburbs for quite some time. What this means is that within the Seattle region more poor people live outside the city than is typical of the average American city. It is from that knowledge that I would hypothesize that the correlation may be negative in Seattle

        Of course, all that depends on whether Salt Lake’s geographic wealth distribution is similar to Seattle or more typical of an average American city. However, if it is the later the findings in Utah may not apply to King County.

      2. Everyone seems to be focusing on trips to and from downtown Seattle. Let’s assume it’s true that poor people are more likely to live outside of Seattle than inside it. (I don’t think that’s true, but it doesn’t matter right now.) That still doesn’t imply that poor people will make *longer trips* than richer people will. As an example, the distance between Renton TC and Southcenter is about 3-4 miles, so the fare would be $1.50 or $1.75.

        If you look at the maps in the study, you can see that poor people travel shorter distances throughout the extent of the UTA’s service area. In fact, low-income people who live on the fringes of the city actually seem to take shorter trips than low-income people who live closer to downtown.

      3. The bombed-out inner-city stereotype occurred only in some northeastern cities. It never occurred in the West Coast or Rockies except perhaps to a small extent in Okaland and parts of Los Angeles. The economic level of city dwellers and suburbanites are about the same, in spite of some poorer districts. It wasn’t that suburbanites were richer, but that they could get more land per dollar there and voted with their feet, and the cities’ populations declined. Now that has reversed because their children got sick of isolated neighborhoods and freeway exits.

        But even now in Seattle you can’t say “the city is richer than the suburbs” or “the city is poorer than the suburbs”. Instead, the Eastside and Shoreline are pretty well off like most of north Seattle, central Seattle, and western West Seattle. South King County is poorer, like much of south Seattle and eastern West Seattle, and Lake City and Broadview.

      4. On land use, the study authors themselves state in their conclusion: “Finally, we caution that with the growing trend of decentralized poverty across the United States, a follow-up social equity analysis of distance based fares would benefit from the incorporation of estimated and projected population characteristics.”

        However, the more I’ve thought about this the more inclined I am to think that, even without additional data, on aggregate poorer people would take shorter trips than wealthier people. I think that because I suspect poorer people are far more likely to run errands using public transit than wealthier people who will generally use public transit only for commuting.

        With that being said, I still think the broader ramifications of such a policy should be considered in more detail and I still think targeted low income fares are a more direct way of addressing this issue, although low income fares wouldn’t preclude distance based fares.

  16. As a transit historian, I find that a lot of the time, the “old ways” made more sense. I say, “Go back to a zone system!” Within the city limits, Seattle Transit System had 3 zones, north, central, and south. The fare just before the Metro takeover was 25 cents base + 5 cents for each zone line crossed. If we update that for inflation, we get $1.50 base + 25 cents per zone. There was also a 10 cent downtown circulator from Pioneer Square to the Westlake area. Reinstating that at a 50 cent fare could replace the now former FRZ (though the routing probably wouldn’t help the homeless much, but maybe on the other hand, that might be a good thing – the free bus that goes to all the social service places is probably a better way to serve them anyway).

    The suburban Metropolitan Transit Corporation apparently based its fares on municipal boundaries, so that each municipality was a zone. IIRC from my studies, fare between Seattle and Federal Way was approximately $1.00 in the early 1970s. If we multiply the fares from that time era by about 6, that pretty much covers inflation from 1970 to 2014.

    MTC was a private, for-profit corporation, and STS, though a city entity, was a lot closer to self-sufficiency than today’s Metro. Food for thought.

    However, I have one comment on the proposed “as the crow flies” distance based fare system: There is a low-tech solution to figuring out your fare in advance if you lack GPS or Internet access: a scale map and a ruler. If such a system is implemented, perhaps Metro could enclose a cheap ruler of thin plastic or stiff cardboard, demarcated with fare amounts, with every system map.

    As to the person who commented that airlines never charge more for a trip with a layover than for a non-stop flight, I suggest they experiment with the Expedia website. There are numerous examples of higher-priced flights with stops or even transfers as opposed to non-stop flights. It’s just the weirdness of pricing in that industry.

    1. New York City discovered a long time ago that one thing politicians shouldn’t be doing is setting fare policy. As far as overseeing the rest of a transit system, they do OK, but with fares… they tend to underprice them repeatedly until the transit system is running out of money. I suppose the same is true of roads, really — why the heck are most of the Interstates free?

    2. At its inception, Metro had the “honeycomb” system based roughly on NJ Transit’s (TNJ at that time) zonal fare structure. This sounds similar to what you describe MTC’s fare structure, in that each town/municipality is its own zone. However, my understanding is that Metro chose to use distance measurements to determine them. They used the STS zones and “honeycombed” around them. But I want to say that Metro quickly (within a year) instituted maximum fares so that you didn’t end up with silly high fares to the rural hinterlands, which in those days were overwhelmingly poorer than the urban and suburban areas.

      Because the honeycomb was so contentious, Metro developed the simplification of City/County/Intercounty. Intercounty fares were calculated separately, and while I only remember the differential for Snohomish County commuter service (CT set the fares collected, 1.25 when the highest Metro in-county fare was 1.00, 2 zone peak), I understand that the Pierce County services provided by the 151 and 432 had an additional fare attached.

      1. It is my recollection that in the early 1980’s Metro’s 1-zone fare was 50c and the 2-zone fare was 75c. Originally there weren’t any peak fares, and when they were first established, I think it was 60c and 90c.

  17. This is a good idea, but I don’t think it is worth it. You are throwing away simplicity for at best, a fee system that is only marginally more just. As it is, you could easily argue that the bus fee is too high. I drive practically every day, and benefit quite a bit when everyone takes a bus. Traffic is lighter, but I’m still able to do my job really better because my customers and coworkers get to work using transit.

    This system just gets too complicated. People will forget to swipe when they get off. They will clog the bus as they get off, making sure the swiping works properly (this already slows down the entrance some times). Meanwhile, any malfunction in the system will either nickle and dime the people who can least afford it, or force others to call Metro and contest a charge (which I’ve done over being improperly charged in the current system).

    Meanwhile, I’m not convinced this would be more progressive. More and more poor people are moving to the suburbs. Yesterday’s working class went over the hill to downtown; tomorrow’s are coming in from Kent and White Center. These are areas where buses moves fast, and distances are large.

    1. This is the usual reason given for NYC Transit’s permanent retention of flat fares — a subsidy to all the poor people living in the Bronx. I’m not sure it’s really accurate, because a fare system like London’s is set up so that it’s cheaper for anyone who stays away from downtown, and a comparable system in NY would be that it would cheaper for people who live *and work* in the Bronx, but more expensive for those who commute to Manhattan. But that’s the reason usually given.

      1. Queens is weirdly structured so I didn’t want to use it as an example. It’s also full of low-rise single-family houses and car drivers. And it’s also full of graveyards. And it’s got areas with no public transit to speak of. And so on.

    2. That swipe on swipe off system would slow down service even more on the one-door bus they use for Route 50.

  18. I think this is pretty much the exact opposite of what we should do. I get that suburban riders are in some ways being sudsidized by a flat fare, but they also pay a cost in time and convenience. And with those costs, there’s still a strong incentive to drive. Lower per-mile costs for suburban routes increase ridership and support for transit taxes. For the subset of the city’s population that is unduly affected by transit costs, there’s always the option of subsidizing low-income fares, as in the new Metro proposal.

    The big gain is simplicity. I’d go even further and get rid of zones and just charge a flat $5 fare, with low-income subsidies. Buying an ORCA would give you a free day’s fare, instantly creating a daily pass for tourists (who could keep their cards for future visits or pass them on to someone else, or just buy a new card next time). Charge ten bucks for a 3-day pass, $25 for a week, and $100 for a month. Covers all transfers/connections on all modes.

    1. Also, this method would eliminate the problems with renewing monthly passes, switching them on and off, and so on. You’d just charge the daily fare every time someone tapped the first time on any day, unless they’d already exceeded the cost of the rolling day/week/mongh pass, which would be free.

  19. “There are other costs, such as training drivers”

    No, just no. Driver training should be simple: “You no longer have to be concerned about fare. Your job is to keep the bus moving – safely.” To put it another way, you wouldn’t pay an accountant $130/hr to accept payment of $.75-$3.25 and answer detailed questions from customers, but that’s effectively what you’re doing when you ask me to stop the bus to perform those tasks. (I don’t make $130, but I and my bus are effectively ‘billed out’ at that rate) Obviously, there is a balance – some questions are quick to answer while passengers with more detailed questions can be referred to other resources. It would also be even more expensive to have a 2nd person always available to help people with disabilities board so that duty will likely aways fall to the driver.

    All that said, with read door readers, a proof of payment system, and fare enforcement officers cross trained in assisting customers (the 2nd “S” in Metro’s Safety, Service, Schedule) something like this *could* work but the goal should be removing complexity from the driver’s world.

    1. I think I was unclear. Presumably, *any* change to the fare system — even if you just eliminated fares completely — requires telling drivers that something is different than before. I didn’t want to gloss over that. But we’re in 100% agreement that the ideal situation is for drivers to completely ignore fare collection.

      Universal proof-of-payment should be our goal. It works for street parking, and frankly, the geographical scope of parking enforcement is much wider than the geographical scope of transit fare enforcement.

  20. What happens to buses that run in loops, i.e. I get on the 47 at Bellevue / Olive, stay on downtown, and get off at the same place — will drivers have to ask riders to reboard?

    What happens if your card doesn’t have enough at the end of your trip (overslept and missed your original stop, e.g.) — will ORCA charge interest on what you owed and couldn’t pay?

    I’m also slightly confused about how transfers would work, based on your answer to Eric about end-to-end trip distance. Do you pay the $1.00 “meter drop” only once per some amount of time, and then charge based on farthest transfer point within the interval?

    e.g. If I take the 545 across the lake and then travel east into the CD, I would pay (Bellevue, Westlake), not (Bellevue, 24th and Olive)? How does this play with incentivizing use / acceptance of a griddy, transfer-based system?

    1. Do you pay the $1.00 “meter drop” only once per some amount of time, and then charge based on farthest transfer point within the interval?

      The ORCA system will attempt to figure out your starting point and your endpoint, and it will charge you based on the distance of your complete linked trip.

      If you end up going out of your way — such as heading to Westlake when you only want to go to the CD — then the cost of your trip will be lower. You probably didn’t really want to go to Westlake, after all; it’s just that the bus network we have doesn’t allow you to make a more direct trip.

      There are a few edge cases. You’ve touched on a couple of them:

      What happens to buses that run in loops, i.e. I get on the 47 at Bellevue / Olive, stay on downtown, and get off at the same place — will drivers have to ask riders to reboard?

      If you take a trip where you effectively travel zero miles, then you pay only the “drop”. Presumably, most people aren’t going to be making trips like this, and so Metro won’t be losing much money because of them.

      There’s a related issue, which is people who exit, make a very quick stop, and immediately get back on the bus, such that their round-trip looks like a single trip with 0 miles traveled. My instinct is that you can be smart about this. Let’s say that someone boards in the U-District, exits at Westlake without tapping off, runs into Kress for 2 minutes, then heads back downstairs and boards a bus heading back to the U-District. They’ll have to tap on when they board, and since they’re taking the same route as before but in the opposite direction, the fare system should be able to figure out that it’s a return trip, not a transfer. And anyway, the average person doesn’t take a long bus ride for a 2-minute errand, so I don’t think this is a common scenario.

      What happens if your card doesn’t have enough at the end of your trip (overslept and missed your original stop, e.g.) — will ORCA charge interest on what you owed and couldn’t pay?

      Well, we don’t want a “Charlie on the MTA” situation. :) When you tap off — or if you forget to tap off — your account will go negative. The next time you try to board, it’ll tell you that you don’t have enough money. It won’t charge interest, but you will have to refill your card before you can board again.

      In the event that you make a mistake, and get billed for a trip that’s much too expensive, you can go to customer service and ask them to adjust the fare.

      1. What if it’s not technically the same route? There are zillions of routes that go between downtown and the U-district.and one can easily go from the U-district to downtown on the 73 and and come back from downtown to the U-district on the 71. One could treat the 71 and 73 as a round trip because you are end at the same stop that you started at, but what happens if someone goes downtown on the 71, spends an hour at Pike Place Market, then returns home on the 66? Somewhere, a computer has to decide whether the user is going downtown and back with a transfer just to travel 3 blocks, or whether the user is actually making a round trip.

        Of course, this question is completely silly to a human – of course the user is making a round trip! But how you would program this into a computer without making up a bunch of special-case rules that add needless complexity to the system is not clear.

        And, rest assured, if there is any kind of fare loophole that allows you pay $1 for what Metro expects to cost $5, people will discover it and exploit it and communicate to their friends on Facebook and Twitter how to do the same. (Of course, riders with monthly passes won’t bother).

        It’s reasons like this that make me feel like a distance-based fare system is just too much complication for too little benefit. If we wanted to focus on improve fare structures, we should be thinking more along the lines of bringing back passes for 1 day, 3 days, and one week. We should also aim for some sort of group or family discount during the off-peak hours. If you think short distances are expensive for a single person, imagine doing it as a family. I can recall at least a couple of times when my parents visiting me from out of town asked me which bus to take to get somewhere and I ended up telling them that by the time they pay for 2 adult bus fares, for just a couple dollars more, they could call a cab. For a family of 4 (with no youth fares or passes, assuming a transfer would be no use for the return trip) a one-mile taxi ride is actually cheaper than a one-mile bus ride.

        The family fares Metro had until a couple years ago were a great idea – stuff like that is what we should be bringing back.

    2. I think that joyrides where you get off where you got on should cost nothing. This isn’t really a problem, as not very many people are actually going to do it. And the city should fund housing for the homeless, by the way.

  21. Fare structures are a complex thing. I don’t think there is a right or wrong way to do it.

    One idea I’ve had is to allow fare setting is based on travel time as opposed to vehicle type or trip end. Think of it like a mobile parking meter or a cab meter. You pay a certain amount per hour. Transfer slips in some systems used to work on this principle, but now with technology we can begin to treat all transit riding in this way. We could begin with a flat fare that would be the equivalent for a period of time (say 30 minutes or 60 minute or whatever is appropriate depending on the route) unless a rider taps off. Then we could have a version of the ORCA card that works like a meter. I think there would be a great benefit to people who aren’t working because many transit trips that don’t involve work are for local errands and shopping and clinics and places like that. A rider could ride to the store or to the clinic and back home without paying twice for a fare. Riders wouldn’t be penalized for making transfers because waiting for a vehicle would be “off the meter”.

    1. I think that a time-based system has several bad properties. The biggest one is that passengers pay more for worse service. Express buses would effectively be cheaper than local buses, which makes absolutely no sense. Also, if you get caught in traffic, not only do you miss your meeting or your transfer, but you also pay a higher fare.

      With distance-based fares, Metro has an incentive to improve bus speed and reliability, but passengers aren’t financially penalized for things that are beyond their control.

      Frankly, I don’t even think that time-based fares are good for cabs. I much prefer the for-hire model, where you pay a fixed rate for a trip, and it’s the driver’s responsibility to take the best possible route.

      1. Going with a smart card tracking system along with incorporating GPS on buses would allow for operators to implement all sorts of pricing adjustments, rather than complicate things for the rider and the bus driver at the time of boarding. Consider how people feel better about their trip if they get money back rather than have hidden costs added to their ORCA card nu traveling into another zone. if we switched to an electronic system, we could build in “instant rebates” if a bus is unreasonably stuck in traffic, for example. If we switched to an electronic tracking system, retailers (or doctor’s offices) could “validate” some of the bus trip just like they validate parking; that would give small storefront retailers something to counter the big box free parking opportunities in the suburbs. Finally, it would change the political landscape for transit. Rather than accept slower buses, riders would lobby harder for faster buses – such as more bus-only lanes and faster loading points – because it would save them both time and money. Technology (ORCA is just the “first generation”) is going to change the way we look at fare payment (like One Bus Away has changed how we look at arrival information) and lots more options will become available as it happens.

  22. I have four comments on this:

    First, you cannot use an average cost per passenger mile as reflective of the cost of providing service. A fully loaded 60-foot bus on an uncongested freeway HOV lane can travel 10 miles at a nominal cost. Another bus may travel less than a mile in 10 minutes if delayed due to congestion, traffic lights, and a lot of boarding passengers. Another bus at the tail of its route has only a few people aboard and still miles to go. And a rush-hour only bus with deadheading at both ends must have enormous costs. Why in god’s name does ST offer Tacoma-Seattle express rush hour buses which can only make a single trip, at an artificially reduced price compared to Sounder trains, which while they have high fixed costs, have negligible marginal costs per additional rider.

    Second, while some people may be forced to ride transit, others are making a choice between transit and driving an SOV. The community and society benefit with that rider chooses transit – through less pollution, less road congestion, and less automobile storage capacity required (acres of subsidized parking, whether surface lots or ugly building pedestals.). It’s bullshit to expect transit to pay for itself while roads have all these hidden subsidies that underprice driving, especially the marginal cost of driving once you already down a car and if you have employer paid parking (which we not only permit but effectively tax subsidize.).

    Third: Keep it simple. Seriously. Simplicity makes it easier for everyone. Riders. Occasional riders. Operators. Visitors.

    Finally, fourth: there should be some fare, both to generate revenue and to create a “cost” so that people appropriately value the service. But the fare structure should be as simple as possible, and it should make it easy to develop efficient service via transfers including mode changes, and it should not provide incentives for duplicated service.

    1. In order:

      1. That’s certainly true, and my original draft of the story actually contained a couple of paragraphs talking about this issue. However, there’s also the issue of fare simplicity, as you say. I think it could make sense to have a few different service classes — e.g. charging a higher rate for one-way expresses. But I want to avoid creating a “double whammy”, where certain neighborhoods get stuck with a bus that is both slow and expensive. Distance is an objective measure, and it’s one that is completely unaffected by any service changes that Metro decides to make, which (I think) is an important benefit.

      2. I don’t believe I ever said that transit should pay for itself. My argument has always been that transit costs money, and that we should try to provide the most effective transit network we can for the amount of money we have. If it’s possible to make a change to the fare structure that raises money without losing a lot of riders or causing low-income folks to suffer, then making that change means that we can run more transit service for the same amount of tax dollars. That seems like a good thing.

      3. Yes, simple is good. But sometimes, it can be worth it to introduce some complexity in one part of the system, for benefits elsewhere. As an example, it would be simpler not to have transfer slips, but that system would give people an incentive to prefer one-seat rides over more efficient connecting service. Similarly, I would argue that the current system gives people an incentive to prefer longer trips over shorter trips, and that a better system would not have that incentive.

      4. I agree with everything you’re saying here, but aside from the restatement of your simplicity point, I don’t see anything that contradicts my proposal. Can you clarify? Do you feel that distance-based fares (as the crow flies, from origin to destination) make it hard to develop efficient service via transfers, or that they provide incentives for duplicated service?

      1. My point 2 was more in response to a few comments that said that transit should pay for itself. That’s a totally impractical goal which would only result in minimal transit service as there are an overwhelming number of road subsidies. If the fares were set to pay all transit costs, you would drive away most ridership that had any alternative.

        It is interesting that much of Europe has relatively high transit fares yet still has strong ridership. The reason is that auto ownership costs and gas costs are quite high. Interestingly parking, while paid for, isn’t necessarily much higher than in USA. But ownership and gas costs are.

        My point 4 was to respond to those who said transit should be free. It shouldn’t be free, and the revenue can be meaningful to support operations. But that doesn’t justify making it harder for people to approach transit.

        I don’t think there ought to be a social engineering purpose of forcing people closer to the city in transit fares. I do think that we should encourage density within the areas that are convenient to transit. One of the things that is interesting in Europe is that often in the suburbs of cities there is a great deal of open space, and then relatively high density within walking distance of the transit station. Zone fares can be in place so that those traveling further pay more. But it ought not to be a goal of the fare policy to discourage people from living in Kent or Bellevue. Much more important to get those cities to encourage density close to the stations and not overprice the high capacity service. In fact, why should the express bus from Kent or Bellevue cost less than the train or light rail?

      2. One of the reasons that I like distance-based fares is that they make all kinds of short trips cheaper, including short trips that don’t go through the city center. It has nothing to do with living closer to the city. Someone from Shoreline who makes a 2-mile trip along Aurora will get a cheaper fare, just like someone who makes a 2-mile trip along Market St in Kirkland, or someone who makes a 2-mile trip along Rainier Ave in Seattle.

        I don’t think it’s “social engineering” to have lower fares for trips that cost less to provide. If anything, you could argue that a flat fare is “social engineering”, since it subsidizes longer trips at the expense of shorter ones. Similarly, you could argue that the current fare structure strongly encourages people who live in Kent or Bellevue to take transit to the city, rather than to commercial centers closer to their home, because the fare is the same in both cases even though the former trip is much further. Right now, you’ll pay the same fare to ride 10 blocks on the B Line as you will to go from Redmond TC to downtown Seattle. Does that seem like a policy that will encourage density close to suburban transit services?

      3. Do you really believe that people would ride a bus from Redmond to Seattle if they could find the same thing 10 blocks away in Redmond, simply because the fare is the same? A round trip to Seattle takes over an hour, and if they try doing it repeatedly they will get tired of it after a few days and will start thinking about what else they’d rather do with the time.

        The reason people make a long trip to Seattle is that they’re going to something that has no equivalent in their city. Or the local transit in their city is so bad that they’d have to walk a long distance to it, or take an hourly bus, or wait 20 minutes for a transfer, so it’s more convenient to go to Seattle.

      4. Do you really believe that people would ride a bus from Redmond to Seattle if they could find the same thing 10 blocks away in Redmond, simply because the fare is the same?

        No. But there are many people who independently make travel decisions.

        Let’s say that Alice and Bob both work at Microsoft, and each one is planning on going to a restaurant after work. Alice is going to Belltown, and Bob is going to Redmond Town Center. What form of transportation will each of them take? There are many factors that go into this decision; one of them is price.

        Suppose that there’s a relatively flat fare, as we have today. Then for Alice, taking the bus downtown is a steal. But for Bob, it’s actually pretty expensive, and he might decide to drive for that reason alone.

        Now, suppose that the fare is based on distance. Compared to the first scenario, Alice is less likely (by some undetermined amount) to take the bus, and Bob is more likely (by some undetermined amount) to take the bus.

        There are thousands of Alices and Bobs, each making independent decisions. If we changed to distance-based fares, the cumulative effect would be that our buses would have more Bobs than they have today, and fewer Alices.

      5. By my values, I do want to subsidize the rider who will take the bus from Kent or Redmond to Seattle more than the one who will take it 10 blocks. First of all the one who will take it 10 blocks has walking as an option, so I’m not concerned about income level. Second, I think a vibrant urban core in Seattle has social value, and I want people from Redmond and Kent to come to Seattle for cultural institutions, restaurants, nightlife, etc. Third, I want to encourage them to ride rather than drive (less pollution, congestion, etc.) Frankly, unless they have a monthly pass, two people coming into town for the evening are going to spend ~$10 on round trip transit fares at today’s prices already. If you raise that to $20, anyone with access to a car will drive as the marginal cost including parking is lower. Finally, if they stay in Redmond or Kent instead, it won’t be because of a cheaper bus fare – instead, they will drive and use free parking provided everywhere in Redmond and Kent.

      6. “for Alice, taking the bus downtown is a steal”

        How much should it cost to go from downtown Redmond to downtown Seattle? And while we’re at it, how long should it take? How many Eastside cities should have a separate all-day express to downtown?

        My instinct is that it should not cost more than $4 or $5 to cover most of the metropolitan area using the trunk routes; e.g., Redmond to SeaTac or Redmond to Tacoma. Currently that’s $3.50 by bus or $4.75 with Sounder.

        “for Bob, it’s actually pretty expensive, and he might decide to drive for that reason alone.”

        That’s actually the same for me, although since I don’t drive it’s “Shall I waste $2.50 or walk?”

        “the one who will take it 10 blocks has walking as an option”

        We’re exaggerating slightly with the 10 blocks. A more realistic case is 20 or 30 blocks, where walking is at the margin of feasability and you may not have time to walk even if you can. Also, I’m not thinking about just somewhere down the RapidRide B, but places like Old Redmond Road or 124th where buses are less frequent or you may have to transfer. In contrast, going downtown and transferring to Capitol Hill or Queen Anne is a piece of cake. It may take longer but it’s just “easier”, partly because all the buses involved are more frequent.

        “I want people from Redmond and Kent to come to Seattle for cultural institutions, restaurants, nightlife, etc… if they stay in Redmond or Kent instead, it won’t be because of a cheaper bus fare – instead, they will drive and use free parking provided everywhere in Redmond and Kent.”

        I want them to have robust access to Seattle too. But I also want better urban villages on the Eastside and South King County, so that those places can have more cultural things accessible from good bus stops.

  23. I would like to see something along the lines of what San Diego did, but with our own tweaks of course. Each route is assigned a designation: Local, Local Express, Urban, Urban Express, Suburban Express and Rural Express.
    Each Designation has its own fare, based on the type of serve offered. 215 & 143 for example are Rural Expresses and should have the highest one-way fare, lets say $5. Suburban Expresses are your 216, 218, 301, 311 etc, and should be $4. Locals are the “In City” routes, like trolley routes, 346, 347, routes that mostly serve local areas, $1.50. A Local Express would carry a surcharge for the premium being offered over its local counter part, same for Urban Express Routes. Urban Routes are like the 5, RapidRides, 101, 271 etc. So depending on where it goes, and distance covered, the Route itself has a fare. If you ride to the end, or think you can ride to Eastgate on the $3 212 to avoid paying $4 for the 218, then you’ll be mistaken.

    To help simplify things for all, the “Route Fare” is charged to all cash paying riders, ALL of them, every time they board! Discounts offered are only available on an ORCA card. Want the Youth Fare, get a Youth Card and load it. Want Disabled or Senior Fares, get the ORCA Card and load it (and there should be no charge for S/D cards themselves). Transfers, ORCA only.

    A sign in the front window, as well as all schedule information will contain the fare for that route. I do not think its cool we charge someone going from Northgate TC to 165th St $3 peak fare, but 342 riders go for $2.50 from Renton to Bothell. Pricing the route makes it more fair based on where that route is going. I don’t think its a perfect solution, but its one worth starting with.

  24. How about zones, the price hoes up by how many zones you travel? King Co. Metro has zones during peak periods so they could do the zone on weekdays until 7 pm time, after that time and on weekends have only one zone. Most Transit companies do have similar idea, they work good and help fund Transit.

  25. Enforcing ride-specific distance based fares would be impossible in a urban transit setting. Much like how the Zone based system is honor based on a busy bus (and the RFA/Pay as you leave was a total joke). As for having fares for routes set on their distance, we can look at Greyhound for an example. Taking the “Web Only Fare” for Tacoma-Seattle which is a duplicate of route 594 from TDS to Downtown Seattle, we find that ST Express is $3.50 for the trip. Greyhound’s lowest price is $8.50. There’s really no way Greyhound can compete in this market since the ST Trip is heavily subsidized, and there are many more trips per day.

    I think a better way to structure fares, would be by type of service. For example, I know Local buses are going to be $2.50 at all times of the day for any kind of trip (electronic transfers only). I also know, that express buses regardless of trip (with electronic transfers only) are going to be $5.00. For my $5.00 I am going to receive a higher quality of service, quicker, more direct trip, better seats, A/C, maybe even Wi-Fi.

    Is this fair to everyone? No. Because not all “Express” routes have a shadow local service. Because some trips may be for 1 stop, vs. someone going all the way from terminal to terminal. Does it make more sense that various Zones and is it easier for the operator to make sure proper fare is collected? Yes. The revenue you may loose on some trips could easily be regained on others since there would be no room for fare evasion.

    Each method has its advantages, on longer trips having a prorated fare makes sense for the operator and the rider. on short trips it does not and simply adds a layer of complexity we don’t need to add to our transit system. As for the obvious questions why cant ORCA handle it, if you look at the existing distance based setups, its Station to Station, where the equipment is located at the stations. Without some reprogramming of the system and more than likely additional hardware (rear door readers and maybe GPS?) could that feature even be used. And only than if EVERYONE has an orca card. Cash fares could simply slip out the rear doors and down the road and there goes your revenue. Not to mention the cost-benefit analysis of making all those system changes, and getting everyone onboard with them and so on so forth…

    Where a distance based fare would actually be good (but is probably illegal) is your ADA paratransit service. We all know this type of service is causing financial hardships on our transit agencies due to its high cost of operation. Charging riders who go clear across the service area for shopping trips and the like a distance based fare would put an end to some of the nonsense and help bring costs down, along with revenue in. While I feel sorry for those who are disabled and low income, truth of the matter is that I have to pay like everyone else, and they are getting a very premium, and customized service at a very bargain rate. Quite honestly its not fare to everyone else who is stuck on some bus that does not run frequently ,and is slow and over crowded with some undesirable people. Meanwhile, these people get a virtually private taxi ride across town or wherever they want to go for less than a buck. But enough of my own opinions…

  26. I think it’s a horrible idea. From a bus driver’s stand point, this will cause so much confusion and even more fare related disputes. Have any of you ever driven the 358 leaving Aurora Village during peak hour and tried to get passengers to pay the $3 fare to cross N 145th St? Everyone either plays dumb and tells you they never knew there was two zones, or they say they’re getting off in Shoreline, then ride past 145th anyway. Or better yet, they pay the extra 50 cents, but give you a hard time, and you get remarks like, “oh, your one those kinda drivers” You get to the point, that you get so frustrated, you just stop asking. Or you get passengers that only pay one zone, who are actually only going one zone, but get upset with you for asking how far they’re going, because they get tired of driver’s asking about it, when they are actually paying the correct fare.

    There is a reason routes like the 5, 77, 113, 120, 124, 131, 132, 303, 306, 312, 316, 358 for example, all cross the zone line during peak hours, but since a large number of the passengers de-board before the zone line, the ORCA on the bus is defaulted to one zone fare to speed up boarding downtown. Imagine if every person boarding the 358 in the PM rush hour, that would get off at or before N 145th St had to ask the driver to override the fare to one zone. But because of this practice, everyone riding two zones actually only pays one zone. It’s to speed up the boarding process and it actually reduces the chances of fare disputes. It’s so much easier this way.

    Also, what about those how ride routes such as 5, 345, 346, 347, 348 for example and only ride a couple stops but, just happens to cross N 145th. Why should they pay $3 even if it’s only a short trip? That’s why I think everything should one fare. Simplify everything. Have one fare $2.25 for off-peak, and split the difference for peak. Say $2.75 for everyone. Raising the fare to $5 or more for long commuter trips, like mentioned above is only going to send commuters back to their cars.

    All I know is when Sound Transit went from having one, two, and three zones, to an easier county or multi-county system, things got better. As for us Metro drivers, driving Sound Transit routes within King County, it is now all one fare. No more having to harass passengers on the 522 or the 550, for example, whether they were riding past Lake City, or crossing Lake Washington to see if they were paying the correct fare or getting off before the zone line. One fare is so much nicer.

    Okay, I wrote too much, but wanted to share what multiple bus fares is like from a bus driver’s point of view.

    1. Your principal argument seems to be that it would be a huge burden to require drivers to enforce this system. I agree. I think that doing so would be a disaster. If we did implement this, it would need to be in the context of full proof-of-payment, where bus drivers don’t have to worry about fare enforcement in any way.

      1. Wouldn’t the fare enforcement costs exceed any potential revenue to be gained? Many cities that have a POP fare policy require boarding passengers to show their payment media to the driver on boarding, at least during off-peak times, because they cannot otherwise cost-effectively enforce POP.

      2. Throughout the City of Seattle, parking is proof-by-payment. There is nothing to stop you from violating parking laws willy-nilly, except for your own conscience, and the roving police officers who enforce traffic policy.

        I’m showing my ignorance here, but if it’s cost-effective to enforce parking policy that way, I don’t see why it wouldn’t be cost-effective to enforce fare policy that way, too. You just set fines high enough to fully pay for the enforcement.

      3. With Seattle’s metered parking, by and large, in one place, I’m sure it would be much cheaper to regularly patrol the city’s parking spaces than to do the same for every bus route.

        Another big difference is that with parking, you have weapons available like the threat of a boot a tow to get people to pay the fines that they owe. By contrast, what leverage does Metro have available to collect a $100 fine from someone who refuses to pay? Especially if that person in question is a tourist who live out of state?

      4. “By contrast, what leverage does Metro have available to collect a $100 fine from someone who refuses to pay? Especially if that person in question is a tourist who live out of state?”

        I believe unpaid fines are usually sent off to collections agencies.

      5. I think that parking meter enforcement is dramatically more efficient and effective than POP fare enforcement. Generally parking meters are in force during hours of peak demand (e.g. 8-6 or 8-8), and the enforcement officer can go from spot to spot continuously, likely checking over 100 vehicles/hour – today their vehicles are equipped with cameras and automatic plate recognition. Second, they have pretty powerful collection resources, which include the ability to impound or boot the vehicle, and to block registration renewal. Officers usually work alone.

        In contrast our buses operate from before 5am to after 1am. Officers generally work in teams of at least 2 – on Link they are generally in teams of 3. The process is inherently more confrontational than leaving a ticket on the windshield. Once the passengers on one vehicle have been checked, the officers would then have to wait for the next vehicle to check, which evenings and weekends may be 30 minutes later. It’s inherently not very efficient. Added to that we have a culture where you get a warning instead of a ticket – at least the first time or two that you are caught – and I don’t think it is an automatic fine, you can go to a judge or commissioner and attempt to make a plea. And then what is the collection enforcement? Even if it is sent to a collection agency, they don’t have the same tools as a parking ticket has.

        I don’t think the POP enforcement on Link pays its costs from fines; it is there to get most to pay. The efficiency on buses must be dramatically worse.

        The premise of your post was to raise more funds for transit. It doesn’t meet that goal if enforcement costs exceed the revenue generated.

      6. Fare evasion usually results in a 124.00 fine. By state statute, any transit system in the state is empowered to institute this fine, which is collected by the district court.

        Failing to pay the fine results in the same blocks as failing to pay a speeding ticket, you get your license suspended. Transit can also suspend your riding privileges if you do not pay.

        Fare enforcement officers and other deputized officials are also allowed to skip the civil infraction process and go right to Theft of Service, a misdemeanor with a potential for a 90 day jail sentence

        There are no statutory freebies, they can go full boat on you on the first infraction.

      7. I don’t think in practice there is any fine at all at the first infraction, just a warning. Then if you get a fine the second time, you can probably contest it (vending machine broken, card didn’t scan properly, just past expiration, whatever the reason) or ask for a reduction in the fine, pleading poverty. Transit riders don’t necessarily have or need drivers licenses nor have vehicles to be booted or impounded. It’s not nearly a slam dunk. Further, do the fines come back to Metro or do they go into the general budget or into the court system?

    2. “Everyone either plays dumb and tells you they never knew there was two zones”

      I’m one of “those drivers”. When I remind the passenger of the proper fare, often they have the extra $.25-$.50 in their hand and drop it into the fare box. It gets old but if they want to play, so can I. I’m occasionally backed by fare enforcement which makes it more interesting.

      Rumor has it there will be a crackdown on two zone fares once the E Line starts. I suspect this will be an unmitigated disaster but one that fare enforcement deals with primarily.

      RE: Sound Transit Fares: Absolutely agree + no paper transfers is nice too. I used to “sell” a lot of ORCA cards at BTC by pointing to the TVM and letting them know I was sitting there for a few minutes. Mixing the best of Sound Transit (No paper transfers, simple fare structure) with the best of Metro (RapidRide PoP + Fare enforcement) would be amazing. Within one shakeup I suspect schedules would loosen up dramatically, especially if Metro didn’t require retapping your ORCA card when transferring between Metro routes (a stretch, but a guy can dream, right?)

  27. In my opinion there are only two ways to make distance-based fares a reality. One is to focus on and charge separately for long-distance express routes – and make them have a premium fare for all riders. It makes those routes less useful or not useful for short trips, so routes like the 358 or 150 should not get the premium fare, but long freeway routes could.

    The second is to make ORCA the only fare payment medium and require tap-on tap-off. Even then I think it ought to be a zonal system, based on the zone where you board and the zone where you deboard after your final transfer – so it is based on distance between zones and not on routing, with perhaps a short-trip fare (under 2 miles?) which applies automatically for short trips.

    Making ORCA universal has all sorts of benefits as well as some drawbacks, but it ought to be feasible, especially if the original cost of the ORCA card can be brought down to $1-2, and there can be ORCA reloading stations throughout the network (either machines, which must accept cash as well as credit, or neighborhood retailers who offer reload as a service that attracts clientele much like lottery tickets.

    1. PS: the short-trip fare would also apply for trips that cross zones, so maybe the implementation is that the 1-zone fare also applies for trips under 2 miles that cross zones, or maybe it’s a separate (lower than 1-zone) short trip fare for all short trips.

      In most zonal systems, the monthly pass is for specific zones, not for a “2-zone fare” which might let someone repeatedly tap off and back on, and would mean they need to pay from an ePurse if they travel beyond the zones of their monthly pass.

  28. Aleks, great post, and it sparked a great discussion. I hope you’ll do more guest posts in the future.

    I’ve been out of town and am late to this party, but I want to echo a criticism several others have made: it’s too complicated. I appreciate that a distance-based system is inherently very fair, and provides a lot of good incentives to both Metro and passengers. But, even with smart cards, people want to understand how much their trips will cost. That’s true not only for tourists but also for locals who want to take trips other than their everyday commute trip — exactly the riders we want to see more of. People won’t memorize complex distance-based rules or the distances between all of the neighborhoods in the city, and the result will be anxiety (about trips of unknown cost) and anger (when trips cost more than expected).

    d.p. is also exactly right that it’s unpalatable to ask people to pay more for a long trip when it takes them forever. So I don’t think a distance-based fare system can work at all on very slow service.

    With both of those objections in mind, here’s how I’d do it:

    1. Local, all-day, core-network buses and mixed-traffic streetcars have a flat, low fare that is the same all across the region. I’d be fine with $2, but $1 would be even better if it weren’t for that pesky revenue crisis. You get a transfer with that $2 if you use a smart card but not if you use cash.

    2. Point-to-point express buses that:
    – (a) have local alternative service
    – (b) are truly faster than the local alternative
    charge a flat fare that is more than the local fare and the same for every passenger on a given express route. There could be one express fare as I’ve proposed in the past, or more than one as d.p. suggests above, with truly long commuter routes charging more than short ones. My thought is that in-city expresses would charge $3 and regional expresses covering more than 12 miles or so would charge $5. (In my ideal world, buses would have color LED signs, and sign color would correspond with fare. Local $2 buses might have amber signage; $3 express buses red signage; and $5 express buses green signage.)

    3. Truly fast regional transit such as Link, which serves both kinds of trips equally well, would have a distance-based fare corresponding with the most closely analogous bus fares. In this scenario, trips under about 6 miles would be $2; trips over 12 miles would be $5; and those in between would be $3.

    This system captures some, but not all, of the value of a distance-based system, but I think it’s simpler, fairer, and much more legible and predictable.

    1. I’m a big fan of local and express fares (and premium express, and super premium express, based on distance).

      One concern I have though, is do we want to incentivize riders to take the 80 (a number I’m making up for a future local route running from downtown to Northgate via the U-District), instead of the 41? Which ride has a larger marginal cost for Metro to operate? For purposes of this thought experiment, pretend Link doesn’t exist.

      1. Well, given the current routings, I’d be charging local fare for the 41 because of the local segment between Northgate and Lake City. And I think that’s fine for Metro, because both the “80” (or as Metro is now calling it, the post-cut 73) and the 41 would operate at or close to capacity, so there wouldn’t be much to be gained by shifting riders between them. If there were extra capacity on the “80” then it would make sense to shift more riders onto it and reduce the number of 41 trips.

        The three-tier fare system I outline will certainly have a few imperfections and odd situations. What do you do with the 255? The 271? The 522? The small number of people who board the 101 along MLK? But it’s possible to handle those situations, and it should be done in the name of legibility and fairness to the highest number of users.

  29. I would also point out that people in Seattle enjoy 62% of Metro service, while only paying about 38% of the sales tax generated–North King (Seattle and Shoreline), South King, and East King are roughly equal in sales tax generation.

    So, distance based fares would be unfair to suburban taxpayers, politically unpopular, and wouldn’t raise much money.

    I would suggest looking at charging more for premium express service first.

    1. Thank you for the inconvenient truth. Doubling urban fares over Seattle would make the inequity four times as great – something that wouldn’t surprise me coming from a Seattleite.

  30. How does this work for monthly passes?
    Metro already has multiple zones, why not just create more zones?
    I never understood, why in the last few rate increases, the multi-zone fares were dropped. I would have expected it the other way round. Zones, in comparison to distance, have the advantage that they are more predictable and allow for buying tickets and monthly passes which only cover certain areas. Distance based systems don’t allow any adjustment on the cost of the monthly passes. Most European systems I am familiar with use zones at least for monthly passes…

    1. What costs big bucks is peak service when more drivers and buses are required, shifts are short, and there’s lots of deadheading. There are plenty of empty seats on most routes off-peak so every additional rider is free money. The intrinsic incentives for ridership decrease because people aren’t going to work, don’t want to wait for less-frequent buses, and see less congestion in front of their car. So the flat fare partly compensates for these and boosts ridership. The zone boundary is especially unfair to Shoreline and White Center. Like Happy Hour, it encourages those who can shift their trips to off-peak to do so, which helps Metro’s overcrowding. And it’s a concession to the poor, who can travel at the minimum rate at least part of the time.

      Most long zone-crossing routes have been split, turned over to Sound Transit, are city/suburban routes, are workhorses, or can be seen as two interlined milk runs. (The 101 is an exception, the 106 is two milk runs, the 5 and 132 are city/suburban routes, the 150 is a workhorse, the 348 was split from the 66 (formerly 305), and the 255 has no reasonable alternative.)

  31. It’s a simple concept of distance based fares but not at all related to the expenses of operating the route. It’s cheaper for metro if you ride the c line from west seattle to downtown and transfer as opposed to taking a commute hour only bus due to the extensive deadheading required. Charge more based on the service not the distance.

  32. My conclusion is that there is no one “best” fare scheme. There are several good schemes, some distance-based and some not. But not everything that works well in its home area would work equally well if transplanted to Seattle. It all depends on the city’s size, geography, distribution of people and jobs, density, and the community’s existing expectations. Metro is a kind of in-between size geographically, larger than several city-only systems (SF MUNI, Chicago CTA, NY MTA) that are doing well with flat-rate, but smaller than systems like BART that really have to be distance-based.

    On the whole I think a flat rate would be best for Metro, but charge extra for the longest peak-expresses. Community Transit does it better: flat rate within the county, but tiered rates to UW, downtown, and north-of-Everett-to-Seattle. In other words, don’t charge extra until the express segment reaches 10 miles: Federal Way, Auburn, Kent, Issaquah, Woodinville. That’s the point where it’s really a special service.

    Another good model is Link: the base fare for 5 miles, and 25c for each additional 5 miles. That gives reasonably wide bands with the same fare (downtown-Beacon Hill base fare, all of Rainier Valley 25c more, TIB 25c more), and it smooths out the “quirks” in the network (where you can’t go directly so you have to go an extra mile to transfer). Charging different rates for 1, 2, 3, 4, 5, or 6 miles is just not worth it, and that’s where the “Broadway-to-Chinatown anger” would arise. So we could just take Link’s scheme. (That would imply $2.25 from downtown to Skyway (124th), White Center (125th), Shoreline (175th), and the next tier beyond that.)

    However, the simplicity argument is strong. Metro’s fares are already too complex, especially combined with ST’s fares. They make visitors’ heads spin. Metro has already gone flat off-peak, and it should do more of that, not less. INSTEAD Metro should shift costs to the longest peak-expresses, and work on a longer-term network that puts medium-distance routes more clearly on one side or the other. No matter how you split it, there will be charges of inequity. Are the 15X, 301, and 216 really premium services, or are they just a skip-stop pattern to manage capacity and compensate for peak-hour congestion? (“When buses get full, group people to different destintations onto different buses.”) An express shadowing a theoretical Link route to a major “station” or P&R doesn’t bother me. What bothers me is expresses that go from single-family neighborhoods or obscure P&Rs to downtown: why should those neighborhoods have an extraordinary privilege that most neighborhoods don’t have?

    There’s another issue too, and that’s the public’s existing expectations. Metro is NOT likely to switch to a distance-based fare, so it’s a somewhat useless debtate. People have expectations about where they can go on Metro and how much it will cost. If we were designing a brand-new system, especially in a brand-new city where things like land use and job locations could maybe be structured differently, then maybe we could eliminate a “41 expectation” and a “76 expectation” and a “301 expectation” from the design. But we’re talking about an existing city with expectations that change slowly, and politicans sensitive to single-family concerns on their reelection prospects. We need to stop worrying about borderline cases in the middle and focus on the most egregious cases at the extremes, and bend the network slowly enough to allow expectations to evolve with it.

    1. As usual, thanks for your wise and measured response. :)

      I don’t quite agree with you that it’s a useless debate. I think that Metro’s current fare structure is broken in multiple ways; it’s not simple, *and* it’s not fair, *and* it doesn’t raise enough revenue. In the same way that Bruce’s and David’s posts are helping to develop a greater public appetite for network restructures, I’m hopeful that we can gradually convince people that some sort of fare structure change would be appropriate, even if it doesn’t resemble what I’ve proposed in any way.

    2. I should point out that metro has used a similar concept before by breaking it out into an insane number of zones. It didn’t work very well so they wound up using the system we have today. It’s not perfect and there are changes that could be made but purely assessing fares based on distance would be too complicated and expensive. Better to charge based on the route. I hope to have my own design which will be based on current routes that exist.

      1. Well, the big difference is that we now have widespread GPS (for determining location) and smart cards (for automatic payment). That makes it easier to implement fare systems that are simple in theory, but difficult in practice. Of course, opinions differ on whether distance-based fares are simple in theory, but I would argue that they’re easier to understand than zone-based fares, if you don’t already know how zones work.

  33. I had the wonderful experience of visiting Japan in 1980. They had this “feature” on their buses. It looked like a big Keno board, with the prices changing based on distance, not time. You could also make change at the front of their buses, and one would only hear a “ding” once per stop, for the system limited it to once. Metro did not have any of these items, most notably the last one, and on yesterday’s #358 (the #359 or the biggest milk run, the #6), you’d be driven crazy by all of the dings. It was several years later that Metro got a clue.

    My reaction to “the board” was intimidation, confusion, and what if I didn’t have enough money? Plus, being a foreigner magnified my concern by at least a hundred.

    Placed into this conversation, the “KISS” (Keep It Simple, Stupid) concept applies. At the King County Regional Transit Committee meeting earlier this month, King County Councilmember Rod Dembowski asked about Metro’s two-zone system. He seemed reluctant to do so, offering a cautious laugh when asking the question of Metro GM Kevin Desmond, who interjected about the peak fare system, which Mr Dembowski didn’t seem to be asking about and which does make sense from encouraging people who have the option, e.g. retirees, to ride the buses midday (I saw this phenomena occur after Metro added a fare differential for senior/disabled fares). Representing North King County, he seemed to be referencing the 3 miles of Shoreline that gets stuck with a two-zone fare, even though it makes not sense and probably costs more to administer, e.g. from the driver having to manually reset the machines to customers forgetting to ask. Eventually, it came out that Metro used to have dozens of fare zones.

    Fares need to be simpler. Ideally, we’d have a regional fare-setting authority (along with other consolidated aspects of transit system management). But, with what we have, Metro might consider a single zone and a wider gap between “peak” and “off peak” fares.

Comments are closed.