are free fares realistic? it depends on the alternatives

In response to my post on Tallinn, Estonia's experiment in free transit for all city residents, a freelance reporter asked me:
The idea I'm most interested in exploring from your post is your proposal that smart farecard systems can be used to easily subsidize fares and "opening up a huge range of subsidy possibilities for any entity that sees an advantage in doing so." I'd like to get more of a sense of what you mean by that and whether this is possible even in today's austerity-obsessed environment.  
What I mean is that as long as the transit agency sets a price for an unlimited ride pass — with appropriate discounts for bulk purchasers — anyone can buy those passes for anyone.  Universities can buy them for their students, companies for their employees, and as in Tallinn, cities can even buy them for their citizens.  Any other entity can also buy them for any group of people it cares about, yielding possibilities that we can barely envision now.
Is this realistic in an age of austerity?  It depends on what the alternatives are.  The alternatives may include building wildly expensive parking, or losing out in a competition for the best people.  

Urban universities with constrained sites often subsidize transit because without it they would need unmanageable amounts of parking.  One common reason that universities get into transit subsidies is that they want to build on their surface parking lots, and the cost of structured parking (and its impacts) turns out to be higher than the cost of buying transit passes for many years.  So it can be a logical business decision.
We're used to the idea that companies leave the cost of commuting to their employees, but companies that are competing for the best talent don't have that luxury.  Witness the huge fleets of shuttles that ferry employees to Silicon Valley giants like Google and Apple from as far away as San Francisco.  Companies that compete for talent can find transit subsidies to be a reasonable part of a total compensation package. And of course, corporate campuses can have expansion crises much like those of universities, where they'd like to build on their parking lots and look for alternatives to expensive structured parking.
City governments are the hardest to imagine financing free fares in the US, if only because of how broke most of them are.  But if it goes well in Tallinn the idea will spread.  One problem in much of America, and notably in California, is that residents are net consumers of government services while employers are net subsidizers of them; this motivates cities to minimize their populations and maximize their employment.  (This explains many odd shapes of city boundaries that seek to include jobs but exclude residents.).  In those distorted tax environments, cities don't want people to live there so much as to work there, so subsidies to residents don't make much sense.  But of course residents are the voters, and wealthy cities that value green credentials may sometimes see merit.  And of course cities aldo benefit if it can reduce their parking requirements, which may increasingly be the nexus that makes fare subsidies make sense.
Remember, though, that massive fare subsidies don't just require the replacement revenue for the fares but also the revenue needed to add service to handle the crowding that the free fares will generate.  I will be interested to see how this plays out in Tallinn.  This has been the barrier to free transit in big cities that have studied it, and the main reason that only small towns — especially university towns — have made large scale fare subsidies work.

14 Responses to are free fares realistic? it depends on the alternatives

  1. Ross Williams March 1, 2013 at 6:04 am #

    As you know, many people in the Portland Oregon already have free bus passes through their employers. Essentially the transit agency has (or had) a program that allows employers to provide free transit for their employees, while paying a fee based on the existing transit to cover the lost fare revenue. Because of air quality issues, large companies were mandated to have a plan to reduce employee use of automobiles. Free transit was a relatively inexpensive way to do that.
    I think it had a big impact on transit use. Unlike many places where transit is used only by the transit dependent, Portland had a lot of people who used it by choice. The free transit made them into transit users. They, in turn, helped demand higher quality service which benefited everyone.
    As you no doubt remember, then Portland Mayor Vera Katz suggested making the whole city fareless. That was resisted by the transit agency. One issue was control over users. Without fares, throwing someone off the bus no longer has much bite. They speculated that there would be an increase in crime and other problems as a result.
    The other issue was the one you suggest, the cost of the additional service. Part of that argument was that, when there is no fare, buses become an alternative to a short walk. All those short trips slow service and inordinately raise costs compared to the benefit. That raised a question of whether that extra cost would better be spent expanding service elsewhere, rather than serving very short trips along frequent service transit corridors.
    BTW, that discussion took place in an environment where many people, including most of the people in the room, already had that alternatives. As always with these discussions, there is an environmental justice overlay of how benefits and burdens are distributed.

  2. Rob March 1, 2013 at 7:00 am #

    Yet another issue is the increased social service role transit needs to serve in a free fare environment. With so many homeless and often deinstitutionalized mentally ill on the streets needing a place to stay warm, a free bus provides shelter and safety (and sometimes a captive audience). That’s not a bad thing necessarily, but it can impact the cost and quality of service – it can affect security and job duties for drivers, as well as drive away some passengers who are not comfortable. It’s just something that has to be factored in.

  3. EngineerScotty March 1, 2013 at 8:36 am #

    Portland Transport did a pair of articles (, about whether TriMet could conceivably become a free service. The latter looks at the Oregon city of Corvallis, a smallish college town (Go Beavs!) with a population of about 80k (including part-time students) and a “pulse” circulator network, that recently switched to fare-free service.
    TriMet has had, for several years, a FRR of about 25%. The big cost driver, as is the case with most US transit agencies, is LIFT paratransit. (An excellent discussion may be had as to why transit agencies, rather than municipal, state, or even federal government, is financially responsible for providing this service, given how expensive it is…) MAX light rail has a FRR of close to one, frequent bus lines are slightly less than that, and the social service bus lines well below 50% (we won’t talk about WES). With recent fare hikes and service cuts, the FRR on mainline transit has doubtless inched up, but new numbers aren’t available yet.
    But if the FRR gets low enough, at some point the cost of collection starts to eat up the bulk of the farebox revenue.

  4. CityBeautiful21 March 1, 2013 at 9:42 am #

    @Engineer Scotty- the FRR data you mention on MAX and Frequent bus/LIFT/etc are very interesting. Do you have a source document you could share on that data?

  5. EngineerScotty March 1, 2013 at 10:24 am #

    It’s a crude estimate. TriMet publishes monthly performance dashboards here:, which give the operating costs for boarding rides.
    As to the revenue per boarding ride–in FY2011, it was almost a dollar (US$96.M of passenger revenue divided by 99.9M boarding rides). Current figures are undoubtedly higher, as fares have since gone up, and the elimination of Fareless Square/Free Rail Zone has caused a drop in ridership, as many “low-value” trips are no longer being taken. TriMet’s full fare is US$2.50 for a one-day ticket and US$5 for a day pass, but significant discounting must be applied to account for transfers, reduced-rate fares and passes, and fare evasion.
    Given that, my initial estimates on per-service FRR (prepared last night before going to bed) may be a bit high.

  6. Andre Lot March 1, 2013 at 2:55 pm #

    Having low, yet meaningful fares, that can be then subsidized by other entities according to their interests is always a preferred solution to free-fare systems, as it allows for the efficiency on the network and individual route metrics to be preserved and taken into consideration.
    Once something is made “free” as in paid by taxpayers and not requiring any payment at point of use, you start having people demanding inefficient routes or service configurations just because it is “their” right. This can go nasty, as in people opposing some route reorganization as they would oppose rearranging a public library system (no one wants THEIR library closed, even it it has extremely low patronage and there is other at equivalent distance that would be expanded).
    Other commentator already mentioned the issue of people riding the system for the sake of it, like homeless people getting free shelter or bored teens joyriding (and inconveniencing everybody else).
    Over the long term, “free” transit could induce other land use problems. Once it is seen as a “right” (instead of a paid service), you could have people requesting routes that serve them directly (regardless of overall network efficiency), of new developments requiring new transit to be served there.
    I bet the net result of any major city adopting “free” transit would be a enormous resistance to high-capacity/higher-speed service such as subways or heavy rail in favor or a messy bus network that is “fair” to everyone by going everywhere, no matter if that means average travel time (or even median travel time) goes up.

  7. LX... March 1, 2013 at 5:13 pm #

    For many smaller to medium sized new world cities with existing sparse low frequency coverage type bus networks the question should be to consider the trade off between making an existing poor transit product free, or instead investing the same money into say a more attractive transit product. For example improving bus frequency, more direct routes, transit priority, better quality stops and terminal facilities, real time information etc.
    Price is only one factor to influence transit use.
    Making a poor product free may increase use by existing patrons but I wonder if the effect on significantly attracting new patrons is as great as some imagine.
    Free isn’t much help if the bus still doesn’t go where you want to go, when you want to go, in a timely manner ,or with shelters and stations that make you feels safe and keep you out of the rain/snow/sun etc.
    My own subjective view of the free transit in Hasselt Belgium when I visited 10 years ago was that it was teeming with teenagers and seniors but probably not much more used by other groups despite being free. But then driving and cycling are pretty easy in a small city of that size.

  8. Matthew March 1, 2013 at 5:38 pm #

    One issue with this proposal using smartcards/passes is that you don’t get one of the significant upsides of free fares: elimination of fare-box equipment and easy all-door boarding. That all has to be maintained for the few who do not possess a free pass. And it may not be worthwhile to keep it around for just them… depending on percentages.

  9. Andre Lot March 2, 2013 at 2:40 am #

    @Matthew: all-door boarding and proof-of-payment are perfectly compatible and not that all expensive and do not slow service down on their own.
    It is also possible to fit stops with touch-based validation machines people tap their cards to travel free.
    Having a no-barrier system is really bad, because it eliminates data collection on users (essential for planning), and makes measures like public transport ban by courts (for those who vandalize the system, for instance) impossible to enforce.

  10. free transit March 2, 2013 at 11:28 pm #

    Why is it that parking lots are a waste of money only on college campuses?

  11. free transit March 2, 2013 at 11:34 pm #

    Studies in NYC showed that if crosstown buses were fare-free, they would move faster and the increased ridership would be handled with the same amount of buses.
    Here are some other places that have free transit besides Tallinn.

  12. Jack Horner March 3, 2013 at 3:31 pm #

    Economic argument: For a non-rival public good like a half empty bus*, making it free at the point of use will increase total community welfare, since extra users can be admitted, and gain benefit, for a marginal cost of zero.
    How much a zero price increases community welfare depends on the details of elasticity of demand. If elasticity is zero (all the people who would use the service at any price or none are already using it), making it free is simply a transfer to those people, and does not change total community welfare.
    The higher the elasticity (the more new users are attracted), the greater the benefit of a zero price.
    Of course taxpayers (or the university administration etc) are still picking up the fixed cost of running the service. Whether this is worthwhile is a different question. The comment assumes that for whatever reason there has already been a political decision to provide the service.**
    Political argument: where transit is free at the point of use, there is no incentive for a public authority operator to improve the service.
    Indeed, more users may mean more costs (when you get to the point where the half empty bus becomes full, and there is demand to provide another one).
    Free transit is an incentive to the operator to run no more than the smallest possible politically demanded ‘charity’ service.
    IMHO, in developed nation cities where the current role and status of transit is weak, yet better transit has good potential to improve the city’s economy and amenity, the political argument against free transit is more important than the economic argument in favour.
    Transit advocates want to be in a space where they can say to government, ‘If you improve the service, it will attract more riders and so *reduce* the net subsidy required.’
    * ‘public good’, in economist-speak, means ‘something that is non-excludable and/or non-rival in consumption’. Traditional example: street lighting. There are no moral connotations. It does not mean (as you sometimes hear) ‘something worthy whose consumption should be encouraged’.
    ** There is a good explanation of all this in the report at (especially paragraphs 5.15ff & 5.32ff)

  13. Andre Lot March 4, 2013 at 4:04 am #

    @free transit: that argument is misplaced. It assumes that front-door, driver-supervised boarding with individual validation is the only possible way to collect fare on buses. I could point to at least 10 big cities where fare collection is done with proof-of-payment and all-doors boarding, sometimes with multiple touch-based validators on-board, sometimes with validators on stops, sometimes relying on a system where 3/4 or more of users have monthly passes they just carry with them.
    @Jack Horner: the marginal cost of additional riders is set to zero (or almost zero) only to the extent the current service can take in more passengers without requiring more vehicles/routes in place.
    When you have “charity-based networks” in big cities, e.g., systems that operate a skeleton service aiming to provide last-resort mobility for citizens who can’t afford other ways to travel, the quality of service might be bad enough that it will not attract a big number of passengers even if the service is free.

  14. EngineerScotty March 4, 2013 at 8:58 am #

    More on TriMet: This recent Portland Transport post contains more accurate information (having had time to properly research the matter).
    And it should be noted that there are different things which one can use as the denominator for computations of farebox recovery ratio. Many transit agencies book “operating costs” that are sufficiently divorced from provision of service (though would vanish were the agency to vanish as well), that they may be excluded from FRR calculations.