🎟️ Eight ways to charge fares for public transport, and what they do
There are lots of ways to charge people for using public transport, and what will work for you depends on what you want to achieve
Good day my good friend.
The next couple of weeks are going to be a bit strange as, rather than writing each post in the week that I post it, today’s and next Friday’s posts are already done and dusted. This is partly as I have a lot of work on over the coming couple of weeks, but also as I need to spend more time researching something I am hoping to publish in a couple of weeks time. In the meantime, enjoy!
This year, Mobility Camp is coming to Sheffield on 30th September, and it promises to be another amazing day of discussion, debate, and taking action. Anyone with an interest in transport should be there, and so should you. You can get tickets here.
The other day, someone who I respect greatly posted something that has started a bit of a debate (as much as debate can be had on social media, anyway). Omer Bor posted his views about the planned change in the bus fare cap in England to £2. His views in short are there are probably better uses for half a billion pounds than a simple nationwide flat fare, which caused a bit of discussion in the comments section of that and one subsequent post - including from yours truly.
Rather than rehash the debate again but in longer form, I thought what would be more useful ns looking at the different pricing mechanisms, and their relative strengths and weaknesses from the perspective of modal shift, equity, and economic efficiency. Each of which is usually contained in the vision and objectives of the transport strategies that such fares are meant to help deliver1.
I also do this partly as I find that people weirdly get really protective about their own chosen method of pricing public transport. Seeing panellists at a conference get quite angry that someone disagrees that free public transport might actually not be a good thing is quite something, I assure you. But hopefully here you may be able to learn something about the value of each method of pricing.
This needs to start from the basis of understanding how public transport use responds to changes in prices, in particular the price elasticity of demand - or how sensitive consumers of a product (in this case using buses) are to price changes. In a synthesis for the Department for Transport, the short-run bus fare elasticity comes out at about -0.4, rising to roughly -0.56 over five to seven years and around -1.0 over twelve to fifteen years. This means a ten per cent fare rise loses you around four per cent of passengers in the first year or two, around six percent over seven years, and ten percent over fifteen years.
That inelasticity cuts two ways, and holding both edges at once is the whole discipline of fares policy. Because demand is relatively inelastic in the short run, raising fares almost always raises revenue, which is why hard-pressed operators reach for it. Because demand becomes relatively elastic in the long run, the same increase quietly hollows out the ridership base over a decade. Everything that follows is a variation on how you distribute that trade-off across different people, times and distances.
Let’s start with the obvious example: free fares. Abolishing fares is the boldest option and the one with the most misleading reputation. The strongest single piece of evidence is Cats, Susilo and Reimal’s 2017 study of Tallinn, the Estonian capital that made public transport free for registered residents in 2013. This work is built on travel diaries from fifteen hundred households before and after the change. It found ridership rose by about fourteen per cent, and the mobility of low-income residents improved, which is a real equity gain. The catch is where the extra trips came from. Much of the growth was newly generated travel and journeys switched from walking, not journeys switched from cars. The modal shift away from the private car, which is the usual justification offered to sceptical taxpayers, was slight.
Dunkerque in France, which went fare-free across its network in 2018, is often cited for a larger effect, with reported ridership growth around sixty per cent and roughly half of new riders said to have come from cars. I would treat that headline more cautiously than the Tallinn figures, because the Dunkerque numbers circulate largely through advocacy and press summaries such as this account from the Observer Research Foundation rather than through an equivalently controlled academic study. Luxembourg became the first country to abolish fares nationally in 2020, and Hasselt in Belgium ran the pioneering scheme from 1997 before abandoning it as costs outran the budget.
The honest summary is the one the UITP reached in its policy brief. Free fares are a blunt instrument. They deliver ridership and a strong political signal, but they are an expensive way to help poorer travellers, because the largest cash saving flows to whoever already travels most, and much of that group is not poor. They also surrender the entire farebox at the moment operators most need revenue to protect frequency, and frequency matters more to most riders than price. A recent review in Transport Policy drawing on more than one hundred and fifty studies reaches much the same verdict, that the ridership effect is real but the car-substitution and equity claims are weaker than advocates suggest.
A flat fare charges everyone the same regardless of distance, and it dominates bus networks in Britain and the United States for obvious reasons. It is cheap to administer, trivial to understand, and reduces boarding times - a major source of delays to buses. This in turn has a secondary effect, because boarding speed affects journey time and journey time affects ridership more strongly than fares do.
The problem is buried in the arithmetic and has been known since Cervero’s foundational work in the early 1980s. Under a flat fare, the passenger travelling two stops subsidises the passenger travelling twenty, and short off-peak trips return far more per mile than long peak ones cost to provide. Because lower-income riders in many cities make shorter trips, a flat fare can be quietly regressive in horizontal terms even where it looks fair on the surface. The 2025 literature review by Kouwenberg and van Oort, which synthesises fifty-eight studies, treats the regressive tendency of flat fares as one of the field’s more settled findings.
There is, however, one small catch here, in that flat fares are only regressive where poorer people take shorter trips. What we know from the National Travel Survey is that, on average, people in households in the lowest income quintile take shorter trips across a range of modes of transport compared to those in households in higher income quintiles. For public transport, the average trip distance for those in the lowest income quintile was eight miles, compared to sixteen miles for the highest income quintile.
But this can vary according to the geographies of deprivation in cities. In cities where low-income households have been pushed to the periphery and now make the longest commutes, a flat fare can protect them and a distance-based one can punish them. The geography of who lives where does more work in this argument than the fare table itself, which is a theme I return to later.
Zonal fares divide an area into concentric or geographic areas and charge according to how many boundaries a trip crosses. London is the obvious example, and the structure is best understood as a deliberate compromise between the simplicity of the flat fare and the fairness of the distance-based one. It captures the broad relationship between distance and cost without demanding the passenger’s exact origin and destination, which historically mattered a great deal when tickets were sold by hand.
The compromise carries its own distortion, the tariff cliff at the zone boundary. Two passengers travelling almost identical distances can pay markedly different amounts if one happens to cross a line and the other does not, and stations or stops just inside a cheaper zone acquire a premium that has nothing to do with the service offered.
Zonal systems also tend to draw their boundaries around a historic centre, which suits radial commuting into that centre and serves orbital trips between suburbs poorly. On efficiency grounds zonal fares are a clear improvement on flat fares and a clear step short of true distance pricing. On equity they inherit a softened version of both parents’ problems, being fairer than a flat fare across long distances but retaining arbitrary unfairness at the margins. For most large networks I think they remain the sensible default, precisely because they are legible to passengers in a way that pure distance pricing is not.
Distance-based charging, or charge strictly by the mile, means that, in principle, price finally matches the cost of provision. This is the structure that performs best in several equity studies. Brown’s analysis of fares in Los Angeles concluded that non-capped distance-based fares combined with time-of-day pricing produced the most equitable outcome once you weighed ability to pay, the benefit riders receive and the cost of serving them. Smartcard and contactless technology has removed most of the historic objection that distance fares were too costly to administer. The reader who taps in and out is already handing the operator the data a distance fare needs.
There are, however, two serious drawbacks. The first is behavioural. Distance pricing directly taxes the long trip, and the long trip is exactly the one where one mode of public transport - the train - competes hardest against the car and where mode shift is most valuable. So a strict distance fare can discourage the journeys you most want to win.
The second is distributional, and this is where you get into all sorts of arguments about fairness. Farber and colleagues’ study of the Utah Transit Authority found that moving from flat to distance fares helped low-income, elderly and non-white riders on average, but that the benefits were unevenly spread. Some residents on the urban fringe were, in fact, made worse off. Where the poor have been priced out to the edge of the city, the fairest-looking structure bills them for the distance that poverty imposed on them. This can be worked around through capping distance-based fares, or grouping locations into fare zones in a mini-zonal fare system. But this is all incredibly fiddly as a workaround. This is a case of a policy that can be vertically progressive and horizontally cruel at the same time.
Anyone who has caught a British train before 9am on a weekday will know all about Peak Pricing. Peak pricing charges more when the network is busiest, and of all the approaches I outline here it has the tidiest economic logic. The cost of carrying a passenger is concentrated in the peak, because it is the peak that dictates how many vehicles, drivers and platforms an operator must own. Off-peak passengers ride largely spare capacity. Charging them the same price is the mispricing that peak fares exist to correct. Furthermore, peak hour travellers - who are often commuters - are often less sensitive to changes in price compared to their off-peak counterparts.
The behavioural evidence is encouraging where it exists, though much of the best of it is imported rather than British. Currie’s 2010 study of Melbourne’s Early Bird scheme, which made pre-seven-o’clock travel into the centre free, shifted about twenty-three per cent of affected peak passengers into the shoulder, a substantial result from a pure price signal. The elasticity work supports the mechanism, since off-peak fare elasticities in Britain run at roughly twice their peak equivalents. Balcombe and colleagues’ 2004 review put peak bus elasticity near -0.26 against off-peak values closer to -0.48, which tells you that off-peak riders are the ones you can move with price and peak riders largely are not. Liu and Charles’s 2013 review of differential fares gathers the international evidence for this peak-spreading effect.
The equity sting is that the people with the least control over their working hours are disproportionately the low-paid. The salaried professional can start at ten and pay the cheaper fare. The cleaner, the carer and the shift worker cannot, and a naïve peak premium taxes precisely the travellers with the least flexibility to avoid it. Peak pricing is efficient, but it is only equitable if it is designed with discounts or exemptions that recognise who is genuinely captive to the peak.
Dynamic pricing has been around in various forms of public transport for decades, though its modern equivalent has largely been imported from low cost airlines. It lets the fare move continuously with demand rather than sitting in fixed peak and off-peak bands. It is the logic of the airline seat applied to the rail seat, and long-distance operators have adopted it enthusiastically. Consultancies describe European high-speed operators such as SNCF extracting revenue premiums of around a quarter through demand-responsive pricing, though I state that this figure comes from industry advisory material rather than independent evaluation, and revenue extracted is not the same as welfare gained.
For the operator the appeal is obvious. Prices track willingness to pay minute by minute, high-demand departures are rationed by price rather than by crush, and cheaper off-peak seats can be sold to travellers who would otherwise not go at all. The problems are informational and political. Passengers cannot plan around a price they cannot predict, and the structure rewards the sophisticated and the flexible over the occasional and the anxious, and the sense that the machine is charging you the most it thinks you will bear corrodes trust in a service that presents itself as a public good.
Dynamic pricing typically suits a competitive intercity market where the alternative is a car or a plane. On a captive urban network, where the passenger has no realistic option and the operator faces no competitive discipline, the same technique reads as extraction.
Yield management is the cousin of dynamic pricing. Where dynamic pricing floats the headline price, yield management allocates a fixed set of seats across fare classes, deciding how many cheap advance tickets to release before protecting the remainder for late-booking, higher-paying travellers. It is the system behind Britain’s advance rail fares.
Used well, it improves the efficiency of a fixed asset. A train has a set number of seats and every empty one at departure is revenue lost forever, so selling early seats cheaply to leisure travellers while reserving capacity for business travellers who book late can raise both revenue and total ridership. The equity picture is mixed. The organised traveller who can commit weeks ahead gets a fare that may be a fraction of the walk-up price, which helps the leisure market, but the person who must travel at short notice, often for reasons entirely outside their control such as a family emergency, faces the full unprotected fare.
The essential caveat, and one the more candid revenue management practitioners concede, is that yield management is a tool for competitive markets. On a route where the operator holds a monopoly, the same optimisation shades from clever capacity management into simple exploitation of captive demand.
The last approach is the one that most of the equity literature quietly points towards, which targeted discounts. Namely, this is keeping a base fare and layer targeted discounts onto it for groups you have a reason to help. Britain’s largest example is the English National Concessionary Travel Scheme, which gives free off-peak bus travel to people over state pension age and to eligible disabled people. The House of Commons Library briefing records that England spent around eight hundred and seventy-seven million pounds on it in the year to March 2023, and that the national scheme sits alongside more generous local schemes such as London’s Freedom Pass.
The Department for Transport’s own evaluation of concessionary travel confirms that free passes generate real additional travel, particularly among the seventy to seventy-nine age group. There is also a live distributional argument, captured in a recent Westminster Hall debate reported by route-one, over whether a universal age-based concession is the right priority when younger people at the start of working life often face the highest fares of anyone.
That does not mean there are not issues. Targeted schemes depend on the reimbursement formula, and the principle that operators should be left no better and no worse off is almost always contested. A former local authority colleague of mine used to joke that behind every discount for bus passengers is a council officer getting an earful from a bus operator, and like any good joke there is a hint of truth there.
Additionally, eligibility drawn by different characteristics that are administratively simple is a crude proxy for need. For instance, why favour discounts for pensioners and job seekers, while missing the working poor entirely?
Even so, targeted measures are usually both more effective and more manageable than blanket free fares when the actual goal is affordability for those who need it.
As you can see, none of these structures win on all three tests. Free fares buy ridership and a political statement at a heavy cost to the farebox and with a weaker equity punch than they promise. Flat fares buy simplicity at the price of a quiet regressiveness that depends entirely on your city’s geography. Zonal fares are the legible compromise most large networks settle on. Distance fares are the fairest in theory and the fiddliest in practice, and cruel to whoever poverty has pushed to the edge of town unless they are capped. Peak pricing is the cleanest efficiency case and the sharpest equity trap. Dynamic pricing and yield management earn their keep where there is competition and a car or a plane to lose the passenger to, and curdle into extraction where the network is captive. Targeted discounts are the imperfect instrument the evidence keeps recommending, provided the reimbursement is honest and the eligibility is drawn around need rather than age.
If there is a single thread, it is the one Peter White’s elasticity work established four decades ago and which the newer equity literature has only sharpened. Fares are a relatively weak tool for moving ridership and a strong tool for distributing cost, so the interesting question is almost never how much to charge in the aggregate but who should bear the charge. The technology to price with real precision now exists in the smartcard reader and in contactless payments. Whether we use it to match price to cost, to match price to ability to pay, or simply to match price to whatever the passenger can be induced to tolerate, is a choice about values. And anyone who says that their fare structure can give you it all is selling you something. Probably their new ticketing system.
While I love writing this, and my writing will always remain free to view, if you are able to donate whatever little you can to help out it is much appreciated. I do have bills to pay after all! You can do that right here.
👩🎓 Further Reading
The clever clogs at our universities, government departments, and other clever people have published the following excellent research. Where you are unable to access the research, email the author – they may give you a copy of the research paper for free.
The demand for public transport: the effects of fares, quality of service, income and car ownership
TL:DR - Probably the most important paper on the subject. It’s a bit old, but it distils decades of British evidence into a single set of demand elasticities, and its headline finding is the uncomfortable one for anyone who thinks price is a lever you can pull hard.
The prospects of fare-free public transport: evidence from Tallinn
TL:DR - The most careful empirical study we have of what actually happens when a city stops charging. Tallinn made public transport free for registered residents in 2013, and the authors tracked fifteen hundred households before and after with travel diaries.
Accessibility, affordability and equity: assessing 'pro-poor' public transport subsidies in Bogotá
TL:DR - Bogotá built a subsidy aimed specifically at poor households rather than a blanket discount, and the authors show it is genuinely progressive, narrowing the accessibility gap between richer and poorer residents rather than simply handing the largest saving to whoever already travels most.
Assessing social equity in distance based transit fares using a model of travel behavior
TL:DR - The paper that stopped me calling distance-based fares simply fair. Modelling travel behaviour across the Salt Lake City region, the authors find distance fares are socially progressive on average, cutting costs for low-income and minority riders while raising them for the affluent. The sting is in the geography.
Equity in transit fare policy: a literature review
TL:DR - It is a systematic review of fifty-eight studies on fare equity, and it sorts the whole messy field into sensible categories running from flat versus distance fares to capping and free travel. A very good overview.
😀 Positive News
Your weekly dose of positivity in a world where bad news dominates. Enjoy.
England freezes regulated rail fares for the first time in thirty years
Announced alongside the Budget in November 2025 and biting at the point the annual rise would normally have landed, the freeze holds all regulated fares, meaning season tickets, peak commuter returns and off-peak returns between major cities, steady until March 2027.
England's national bus fare cap set to fall back to two pounds for 2027
Having held the single-fare cap at three pounds through 2026, the government confirmed in July that it will return to two pounds for the whole of next year as a cost-of-living measure. The figure attached to it is a government estimate that the earlier two-pound cap generated around thirty million additional bus journeys in ten months.
Ho Chi Minh City sees bus ridership jump more than thirty per cent after going fare-free
From 1 July the city removed fares on 134 subsidised routes, and within five days it had recorded more than 1.32 million trips on those routes, according to the city's Public Transport Management Center reported by VnExpress. The first day carried nearly 274,000 passengers, about one hundred and twenty-eight per cent of the same day a year earlier, and ridership peaked near 283,000 on 3 July.
The City of Boston reports that removing fares on routes 23, 28 and 29 enabled all-door boarding that cut the time buses spend stopped at stops by about twenty per cent, with ridership on route 28 recovering to more than ninety per cent of its pre-pandemic level and nearly 23,000 fare-free trips taken every weekday across the three routes.
Massachusetts public transit passes its pre-pandemic peak for the first time since 2019
Statewide ridership has climbed over twenty nine percent to surpass 2019 levels, as reported by the Daily Hampshire Gazette, with the fare-free Pioneer Valley Transit Authority contributing a twenty one percent rise in the 2025 financial year followed by a further seven percent in 2026.
💻 Hard Work
I probably won’t write this section for a couple of weeks due to the reason given in my introduction. So if you really want to hear what I have been up to which isn’t meetings, Teams calls, and riding up and down railways, you are going to have to wait. Sorry.
🎶 Musical Finale
You can hear the guitar riff of Iggy Pop’s The Passenger in your head right now, can’t you? I don’t think there is a more iconic song about being a bus passenger.
🪧 Notices
This section is intended to notify you of important things that may or may not annoy you, depending on your view of the world. But these are things I feel it is worthwhile pointing out.
Regarding Artificial Intelligence. For my newsletter, I use AI like a research assistant. I get the AI (in my case Claude Opus 4.8) to find for me well-cited and evidenced research in the subject matter, focussed on academic journals with a high impact factor. I follow this up with my own research, and I check each source myself. Where I have the time to, I contact domain experts or seek out their published works to validate the findings. For this particular article, I did get AI to write a first draft of some paragraphs, to give some ideas for structuring the article, which I subsequently edited into my own words. Otherwise, the words written here are my own, spelling mistakes and all.
Regarding my expertise. I am a transport planner, so my realm is public policy, how people and things move, and their impacts. While I research everything in a manner I consider robust, there may be some instances where my lack of domain expertise fails me. That’s on me, but that’s all part of learning. I always encourage any reader to do their own research on the subject matter.
✅ Sign Up
If you have been sent this newsletter, please consider signing up. Its free to do so, and I don’t spam your inbox with random rubbish - just my own words.
I could put in another which is to ensure profitability, which is a key consideration in a market where the public transport operators operate services on a commercial basis. But for the purpose of this newsletter, I am going to try and keep things simple.
Good read.
I think the point that flat fares on buses reduce boarding/disembarking times and therefore journey times should not be understated. Bus stop dwell times are a major cumulative source of delay. Every second counts.