The Horrors of “Transferring” in 1974, and a Happier Future

Connections, or transfers as North Americans depressingly call them, are the foundation of a simple, frequent transit network that’s there whenever you need it.  I laid out the basic argument here, but in brief, a transit system that tries to run direct service from everywhere else (so that nobody has to make a connection) ends up as a confusing tangle of hundreds of overlapping lines, few of which are frequent enough to rely on or simple enough to remember. 

But once you decide to offer free connections, you face challenges.  Rail rapid transit systems usually let you change trains behind the faregates without paying another fare, but surface transit systems (bus, rail, or ferry) struggle with how to encourage transfers without encouraging fare evasion.  If you give paying passengers a transfer slip providing a free ride on a connecting service, you get all kinds of abuse: people who don’t need the transfer slip sometimes sell it, or give it away, or simply drop it where someone else can find it, and ride free.

1974transferprotocol_web The blog of Vancouver’s transit agency TransLink, called The Buzzer, recently featured this 1974 explanation of “transfer regulations” in that city, as printed on the route map. It’s a nice example of the kind of regulation that was common at that time.

Note how much complexity follows from trying to insist that transfer slips must be used only for connections and that you must not do anything else while connecting. The authors have thought of everything you could conceivably do that isn’t exactly what they intend, and are resolved to prohibit all of it in meticulous detail.

Fortunately, not many transit agencies take this tone with their customers anymore.

There seem to be three defensible approaches to connection-pricing, at least in the North American and Australasian contexts where I work:

  • One is to stop giving transfer slips, but instead sell a reasonably priced day pass.  This is now common in parts of the US, especially in the sunbelt.  With a day pass, you charge a bit more and in return you give up trying to control how the slip is used.   Of course, this is still unfair to people making spontaneous one-time trips, which are, in my view, a pretty foundational element of healthy urban life.
  • An intermediate approach, modeled by AC Transit in Oakland, is to provide a transfer slip for a nominal extra charge, such as 25 cents.  This is supposed to deter people who don’t need one from taking them, thus reducing the number that are discarded where others can pick them up and use them.
  • The third approach, which I have to say I prefer, is to take a deep breath and just not worry about it. Give out free transfer slips because the design of your system is based on encouraging connections.  Treat the transfer slip as a 90-minute or two-hour pass, good for unlimited rides within that short period.  Do what you can to discourage them from being sold or transferred, but don’t try to control how they’re used by the person to whom they’re issued.  If someone can make a quick round trip in that time, or do a stopover, then that’s fine.  Portland’s TriMet, Vancouver’s TransLink, and San Francisco Muni all do this, as do the integrated networks in Melbourne and Brisbane, Australia.  In all these places, and many more, a base fare buys you unlimited use for 90-120 minutes with the fare zones specified.

Of course, there are lots of other kinds of connection-pricing policy out there, including a range of partial policies and non-policies.  It all gets much more complicated where there are multiple entangled transit operators watching their own balance sheets and no citywide agency with the power to coordinate them.

Many of the security problems around transfer slips should disappear in the new era of smartcards.  A good GPS-based smartcard system will know that you got off another transit vehicle nine minutes ago, so it will be able to determine whether you should be charged a new fare when you board the next one.  It will be interesting to see whether agencies that still have fare penalties for connections take the opportunity to eliminate them.  It will still be politically hard: if you eliminate your connection penalty without raising the base fare, your revenue will tick downward.  If you raise the base fare to compensate, lazy journalists will do stories about your evil fare increase, ignoring the larger fairness of the change.

But one thing is true everywhere, because it’s a matter of geometry: if you want frequency and simplicity in your network, you have to encourage connections.  So any pricing scheme that penalizes connections, or that shrouds them with punitive rules like this 1974 policy, is in conflict with those basic goals.

41 Responses to The Horrors of “Transferring” in 1974, and a Happier Future

  1. anonymouse November 16, 2010 at 1:31 pm #

    Your “good GPS-based smartcard system” sounds really, really scary to me. Basically, like tracking where everyone is at all times. And the real reason behind day passes in LA at least is that the county apportions money among the various local transit operators by “ridership”, but since actual ridership is (or was) hard to measure, “ridership” is defined as revenue divided by base fare. The lower the base fare, the higher the “ridership” and the more county money you get. A low base far and no transfers allows agencies to inflate their “ridership” to get more funding. Really.

  2. Nick November 16, 2010 at 3:28 pm #

    Except you really don’t need to track where everyone is at all times (and it’s not even possible to do anyway, for technical reasons): Just where and when they boarded the previous vehicle, so you can determine (based on the current time and location) if the passenger should be charged another fare. I would welcome such a system if it meant I didn’t have to pay an extra base fare to get to the just-out-of-walking-distance bus route (even if the base fare were higher than it is currently).
    And really, what’s so scary about it? Ignore for a moment the fact the smart card system already knows the time and approximate location of each machine that scans your card: Your cell phone is an even better tracking device.
    Of course, there’s a lot of low-hanging fruit to pick when it comes to fixing LACMTA’s bass-ackwards smart card system.

  3. anonymouse November 16, 2010 at 3:39 pm #

    @Nick it’s not the boarding I’m worried about, it’s the knowing when you got off even if you don’t have to tag off. Read carefully, because there’s a fine line here between something more or less reasonable and Big Brother.

  4. NCarlson November 16, 2010 at 3:47 pm #

    “Your cell phone is an even better tracking device.”
    This. As much as there is some tracking potential in a smartcard system there are a lot of technologies out there just as or more ubiquitous that do a much better job. It’s not even like we’re talking new tech here, I don’t believe anyone is using GPS specifically, but Oyster among others certainly meets this definition of a good smartcard system, and even with all the surveillance going on in the UK no one is particularly concerned about this.

  5. Alon Levy November 16, 2010 at 4:02 pm #

    You don’t need all these regulations when you have a good POP system. Anyone with a valid unlimited card should be able to board buses without trouble anyway. For the rest of the population, the transit agency should encourage the use of reusable tickets, such as Octopus or ez-link; even a magnetic card like MetroCard would do. If the fare is based on distance, then there’s no need to make people tap out during the transfer. The few people who use single-ride tickets should get receipts with the entry time printed clearly.
    The back end of such a system would be quite complex: each bus would need a card validator and potentially a farebox at every door, the smartcard system would need to know what to do with people who tap into two routes without tapping out in between, and the agency would need to keep a stable of fare inspectors. But the front end would be simplified, allowing people to transfer wherever and whenever they want.

  6. Mark November 16, 2010 at 4:15 pm #

    I’m with Jarrett and the “don’t worry about it” philosophy on connections/transfers. Over the Noon hour today I stopped to buy a new external hard drive, go to the bank, and go home for lunch on one outbound bus trip, and then return to work on the same inbound route, all on one two hour transfer. Victoria BC has been using what we think of as the Vancouver transfer system for a few years now and its very flexible, convenient, and easy to use.

  7. Chris B November 16, 2010 at 4:20 pm #

    Montreal has already implemented this with their Opus card:
    “When you charge tickets onto your OPUS card and use them to pay your transit fare, you have 120 minutes from the time you initially board a transit vehicle to use your transfer. If you have to use more than one STM vehicle (métro or bus) to complete a trip, you simply have to hold your OPUS card above the card reader on the bus fare box or the métro turnstile to validate your transfer. If you use an STM vehicle after the 120 minutes allowed for your transfer, another ticket will be deducted from your OPUS card.”

  8. Joseph November 16, 2010 at 5:47 pm #

    @Chris B
    That was initially a bug with OPUS, but then they decided to stick with it, and I sure am glad they did.
    Car and bike drivers can do multiple short errands for the same price as if they didn’t stop, so why shouldn’t transit users?

  9. Aaron November 16, 2010 at 6:30 pm #

    The OPUS system in Montreal is pretty good, but it sucks for transfers that involve leaving and re-entering the metro (subway) system. You’re transfer time limit is applicable to Metro > Bus, Bus > Metro, or multiple bus > bus > bus transfers. But if you take the metro, exit, and then re-enter the metro, even 20 minutes later, you’re charged another ticket.
    Also, Edmonton, Alberta has had a 90-minute unlimited transfer system for years now (2-3 years since I’ve been, however, so this may have changed).

  10. Wai Yip Tung November 16, 2010 at 6:46 pm #

    That smart card system does not have to be GPS based. If it can tell the boarding is within like 2 hours then it is good enough.
    That’s what happened for San Francisco Muni. Their drivers used to causally hand out paper ticket good for 4-5 hours since they bother bother to tear the time limited precisely. Pretty much one ticket will be good for a whole morning or afternoon. This privilege is gone now that a smartcard and record the boarding time correctly 🙁

  11. ant6n November 16, 2010 at 6:47 pm #

    That actually makes sense, coming from what Jarret says: The Opus ticket only allows to enter the metro once, so it’s useful only for one trip, as it is designed to be. It’s not a 2 hour unlimited pass, it’s a ticket for one journey. I am not quite sure, but I was under the impression that you can’t even take the same bus twice (you can’t go back and forth).
    The creative thing to do is to take the metro one way, and the bus in the other. or go in circles. Which is possible given the pretty good bus system in Montreal.
    In Berlin, single tickets used to be 2 hour unlimited passes. At some point it was changed so that tickets cannot be used to go back to the original destination within 2 hours, you are not allowed to travel in circles. Since the city uses POP, this is only enforced via fare inspectors. Although I have a feeling this can lead to all sorts of gray areas – it’s hard to define what a ‘circle’ is. Although I have never heard of any issues.

  12. Nathan Williams November 16, 2010 at 6:49 pm #

    When Boston went to their dual smartcard/magstripe system a few years ago, one of the things they did to try to encourage use of the smartcard option was to make it the only way to get transfers. If you use the magstripe ticket, you just have to pay another fare. The smartcard system allows one transfer, and charges you the fare differential if you transfer from bus to subway. Doesn’t seem terribly complicated, all things considered.

  13. JJJ November 16, 2010 at 8:30 pm #

    The best systems, IMO are the “this ticket is good on all services for 90 minutes” kind.

  14. Chris B November 17, 2010 at 4:13 am #

    @jjj – that brings back my youth when we would try and get to the mall (or wherever) and catch a bus one or two minutes before our transfer expired.

  15. Tom West November 17, 2010 at 7:01 am #

    Time limited transfers pose problem for larger systems. In Toronto (for example), there are potenital trips where the shortest journey time is well over two hours. If you set the transfer limit at 90 or 120 mins, you’d end making people pay twice for some one-way journeys. If you set it much higher, you start getting a significant number of two-way trips resulting in a one-way fare.

  16. Chris November 17, 2010 at 9:20 am #

    I much prefer transfers and lament the fact that few agencies in California offer them anymore. However, it would be interesting to hear about this from a transit operations perspective – I think most of the complaints between driver / passengers involved transfers; if true, do agencies that no longer have them experience more harmonious driver / passenger interaction?

  17. TransitPlannerMunich November 17, 2010 at 9:54 am #

    As probably expected, of course an example from Munich:
    – POP-system
    – most riders have monthly passes
    – tickets for single rides are stamped when entering a vehicle or station at a
    machine and get a time code and code for the stop/line/direction
    – transfers and interruptions of the trip are permitted, but not going back, and it
    is valid for the following time period:
    “Short distance” ticket (4 stops): 1 hour
    City ticket: 3 hours
    Tickets including the suburbs (up to 40 km outside of Munich): 4 hours valid
    You can change so much as you like or need between all buses, trams, subways,
    commuter rail.
    But you cannot go back – unless you buy a day pass of course.
    Simple for the transit riders and the transit authority.

  18. Plattypus1 November 17, 2010 at 10:06 am #

    Of course, to add to the disaster that is transit fare policy in LA, Metro does issue transfers (for a nominal $0.35). Many out-of-town visitors from transfer-using systems will buy a transfer in order to board a second Metro vehicle- but, surprise! Metro transfers are only good on NON-Metro vehicles.
    I get really frustrated with the TAP haters in SoCal, because all of the little balkanized municipal and suburban operators all have their own fare structures and media, and transfers between them tend to be abysmal. We need a single, regional fare media and, if possible, fare structure, and- though it has its flaws- TAP is the best vehicle to accomplish this. It’s certainly better than what we’ve got now.

  19. TransitPlannerMunich November 17, 2010 at 11:02 am #

    The funny thing of the American system is:
    You are punished anyways when you have to change to a different line to reach you destination, including a loss of comfort and time.
    But instead being happy that this person is using public transit despite he/she has to change the vehicle, the person is charged with an extra fee.
    I think no other industry could afford to treat its customers like that. You only do that when you have a monopoly and your clients do not have an alternative.

  20. M1EK November 17, 2010 at 11:27 am #

    TPM, it’s even worse than that – Jarrett and many others are critically underestimating how much of a pain transferring really is. And the idea that great transit systems rely on transfers is misleading. Great transit systems don’t bring everybody to one place, force them all to transfer, and then deliver them to their final destination – they deliver a TON of people via direct routes, and handle some set of the remainder with transfers.
    What modern transit planners seem to be pushing is more of a huge-transfer-barn-on-the-edge-of-downtown approach (and it’s producing easily foreseeable results here in Austin, for instance, where a rail start is struggling to hit 1000 boardings/day because it forces every single passenger to transfer to shuttle-buses).

  21. Joseph November 17, 2010 at 4:05 pm #

    You can read Jarrett’s blog for yourself, but I think his opinion on transferring is very correct.
    Transferring doesn’t have to be a pain. My bus company (suburban) schedules its buses to arrive at the hubs all at the same time and leave 5 minutes later. When a bus is late, the buses people need are held to assure transfers. And the system works! Even in my suburban area, you can travel literally anywhere with a great frequency and reasonable travel times.

  22. M1EK November 17, 2010 at 4:50 pm #

    Where are you from? What’s the cost of gas like there, and do middle-class people typically own automobiles?
    Because there isn’t a city anywhere in the US – not even New York – where a two bus commute with a 5 minute transfer won’t turn off anybody with any choice. Even in the rest of ANZUS, it’s hard to imagine anybody with a choice putting up with that commute unless they have no access to parking at their destination.

  23. Paul C, Vancouver November 17, 2010 at 5:25 pm #

    Its too bad that there isn’t a system that publicly announces who has not paid the proper fare to be on that transit vehicle.
    If someone got on that has not paid and suddenly there was flash lights and arrow pointing to this person. I’m sure public peer pressure and scorn would force this person to pay. The problem is no on really know who has and hasn’t paid the proper fare on a transit vehicle. Thus people get on without any remorse in doing so.

  24. Paul K McGregor November 17, 2010 at 5:34 pm #

    In Dallas, when DART created the suburban system, they did adopt a hub and spoke system with timed transfers and pulses at transit centers. This has worked real well over the years and has carried over to the light rail as well. However, the challenge of keeping the pulses on track is that added congestion starts to slow the buses down and that leads to either adding more buses to a route or forgoing the timed transfer.
    Smart cards can certainly simplify the process. However, using transfers gives some discretion as to the length of time given, either with the cut of the transfer or a driver’s discretion to let you board. With a smartcard, you no longer have that discretion. If you have two hours to travel and you reach two hours and one minute, you get hit with another fare. People do get annoyed with that but yes, we all have to adapt to new ways of doing things. So it should eliminate cheating and hopefully increase revenues as well.
    @M1EK after reading your many comments, I see why you have such a negative view towards transfers. I worked at Capital Metro at the turn of the millenium and at that time, I did have discussions with staff about doing timed transfers. It was not pursued because of the added cost involved. Now this dicussion has spilled over to the train and how people do not want to ride a shuttle for the last mile trip. The bus is there to meet the train so I am not quite sure what the problem people have with this. The overall travel time might be the issue but the notion of transferring should not be.

  25. Alon Levy November 17, 2010 at 6:48 pm #

    M1EK, you should probably not rant about transfers to a person who lives in one of the countries that invented the timed cross-platform transfer.

  26. Joseph November 17, 2010 at 7:16 pm #

    M1EK, I’m from Ste-Therese in the north of Montreal. And, yes, public transit is only used by about 5% of workers travelling locally and about 25% of those going to Montreal but is used by a much larger portion of the student population.
    But, what I’m arguing is that a direct-ride network would be even less productive. It would simply not be possible to provide a decent level of service between every residential area and every center of employment or study at any frequency that people would use. Sure, if the bus were direct and at the right time for their shift, people will take it, but at the frequencies a direct-ride system to entail, that would not be very likely.
    A transfer-based system ensures that everyone has access to public transit for all trips, regardless of destination, at times convenient to them and their shifts/courses. It also means that there are fewer routes so that anybody can literally know the whole system by heart. And because the routes run at fixed hourly intervals outside of rush hour, it is possible to know the entire schedule by heart easily. This is what is needed to get people to use transit for trips other than commuting.

  27. ajedrez November 17, 2010 at 7:28 pm #

    This reminds me of when I lived in Brighton Beach, Brooklyn a few years ago. We had 2 north-south bus lines 10 blocks apart and a subway line right in the middle, meaning that people could take the subway to their destination and walk a couple of blocks and take the bus back.
    I like the “unlimited pass for 2 hours” idea with transfers. First of all, there are many variables to consider when making connections. For example, there is no way of telling if the bus is too crowded to stop, and therefore it bypassed people waiting at the bus stop to transfer. Or maybe some people were quick enough to make the connection and others weren’t.
    Also, I feel that allowing return trips (if they are done on a different mode of transportation) can encourage some people to use mass transit, as they can go to their destination, accomplish a few errands, and come back on one fare.
    Or it can encourage them to spend money while making the transfer-for example, by picking up their lunch at the transfer point and spending 10-15 minutes there.

  28. EngineerScotty November 18, 2010 at 6:27 am #

    @Tom West.
    A TriMet ticket (in Portland) is good for two hours–with the added provision that if you board a vehicle within the two hour window, you are permitted to finish your journey on that vehicle without being in violation (i.e. you don’t have to buy another ticket).
    That said–I can imagine journeys on the system which are >2 hours, and for which one must board the final leg more than 2 hours after getting on initially. (Taking the 57 from Forest Grove to Hillsboro, taking MAX all the way out to Gresham, and then boarding a local bus at the other end to reach a final destination, for instance). But such journeys are probably pretty rare. (For that reason, it might be useful if all-zone tickets were good for 3 hours rather than 2…)

  29. M1EK November 18, 2010 at 6:30 am #

    @Joseph, you’re arguing against what has worked in ANZUS cities with higher modeshares for transit among choice commuters.
    Again, not every route has to be direct, but you also can’t go the currently preferred route of building a big transfer center somewhere where it’s convenient for the transit agency and make nearly 100% of your potential customers transfer. And you certainly can’t ever get to a city where a lot of people are transit-dependent by choice by starting out with a ton of transfers.

  30. M1EK November 18, 2010 at 6:49 am #

    @Alon, I remain very unimpressed by the idea that timed cross-platform transfers are somehow penalty-free. Arguing their effectiveness when they’re deployed mainly in areas where driving is an order of magnitude more expensive and inconvenient than in ANZUS is just ridiculous.

  31. M1EK November 18, 2010 at 6:51 am #

    @Paul, if you don’t understand the problem, then you don’t understand choice commuters, like the rest of the people who thought the trains would be leaving people stranded at stations for lack of room.

  32. TransitPlannerMunich November 18, 2010 at 12:05 pm #

    I understand why you dislike transfers. But indeed a complex public transit
    system cannot operate properly without it.
    It is impossible to offer direct connections for any trip.
    What you can do is to minimize the amount of transferring:
    – when you can reach the center of the city and the center of your district
    without transferring, it is a plus.
    – when you can reach 80% to 90% of the possible destinations with only
    transferring once it is a plus.
    – and when you need to make those transfers it would be vital that they are so
    easy as possible. So without worrying about extra tickets, transfer slips, time
    tables including long waiting times, long ways between the stops.
    So indeed, as Alon Levy mentioned, the ideal transfer stops, for example in
    Munich’s metro system, are stations like Scheidplatz and Innsbrucker Ring. At
    both stations two metro lines meet at the same platform. So when you travel
    southbound with the U2 from Feldmoching or U3 from Moosach both trains arrive at
    the same minute at the station. The doors are opened and whoever likes can walk
    a few meters across the platform to change. One lines runs directly towards the
    central station, the other continues to the heart of downtown (Marienplatz).
    After a few seconds all passengers who changed are on board again and the trains
    continue their journey.
    Also at other transferring points where north-south-lines meet east-west-lines
    the system is designed that the transfer is so short and easy as possible. Without the long walks that you need in older systems like London or Paris, where you
    have sometimes the impression that you walk a longer distance than you travel by

  33. M1EK November 19, 2010 at 6:32 am #

    @TPM, obviously a comprehensive system that covers a non-trivial area requires some transfers. But the idea that you can go into a metro area where transit is currently the purview of the poor, infirm, and elderly and get people to leave their cars at home by delivering a starter line that requires every passenger to transfer to a bus is just ludicrous.
    Yet that’s what we did in Austin – following advice of people who got the wrong message about transfers by looking at cities with mature transit networks that already have huge modeshares among choice commuters.

  34. ant6n November 20, 2010 at 11:43 pm #

    Your argument is not particularly convincing. Joseph tried to outline how a timed transfer in downtown, together with fixed interval schedule allows him to travel almost anywhere easily (without having to look up schedules, too); and that with the downtown timed transfer, the system is vastly simplified. You are arguing (basically) that if he wasn’t hold or poor, then he’d take the car. This is a claim that would need to be backed up.
    At the same time it’s not clear how even less frequent one-seat ride systems (which are also more expensive because fewer potential riders) would be favorable.

  35. M1EK November 22, 2010 at 5:41 am #

    @ant6n, Austin has timed transfers for its commuter rail start right now. It’s seeing pathetic ridership compared to projections for a direct light rail proposal that it supplanted.
    It’s really you guys that have to come up with reasons why facts on the ground don’t prove my point, not the other way around. Again, mature transit-rich areas with large populations of transit-dependent (voluntarily or otherwise) aren’t really relevant to this discussion.

  36. Matt Fisher November 23, 2010 at 3:17 pm #

    Why is it that transfers are so important to be something to avoid in Ottawa? If the Transitway is converted to light rail, everybody will inevitably have to transfer between bus and rail, even if I support LRT.

  37. anonymouse November 23, 2010 at 8:09 pm #

    M1EK: the problem is that you only get one transfer, and with something like a commuter rail, that transfer is going to be from some feeder service (or car, via park-and-ride) to the train. The train then has to take you directly where you want to go (Downtown, for example). Also, I would conjecture that seemingly small details in how the transfer is implemented can make a huge difference in the number of passengers transferring: random vs. timed vs. guaranteed connections, across the street vs. cross platform vs. same platform, that sort of thing. In the end it all comes down to the same thing: travel time and reliability, but it’s passenger travel time that matters here, not the vehicle travel time that most planners look at.

  38. ant6n December 2, 2010 at 5:21 am #

    The Austin system is pretty flawed, right from the beginning “us guys” noted that.

  39. M1EK December 2, 2010 at 9:19 am #

    ant6n, you need to go look at who was the source material for some of the references in that article. Hint: you’re talking to him.
    It was most definitely NOT the case that people knew ahead of time that the requirement to transfer would doom ridership. There was exactly one guy making that case in 2004. Again, you’re talking to him.

  40. Alon Levy December 2, 2010 at 12:06 pm #

    You called the Austin line’s failure ahead of time, and think transfers are untenable and hourly off-peak service is good frequency. Richard Mlynarik called the BART to SFO extension’s failure ahead of time, and thinks transfers are indispensable and hourly off-peak service is as good as no service. Who should I believe?

  41. M1EK December 3, 2010 at 6:16 am #

    Alon, I never said hourly off-peak was “good frequency”, but that for a commute-oriented service it was enough to demonstrate demand (or lack thereof).