A thread for discussions on free or near-free evidence-based PT research.
The impact of fare-free public transport on travel behavior: Evidence from a randomized controlled trial ($)
Regional Science and Urban Economics
Volume 86, January 2021, 103616
Authors: Owen Bull, Juan Carlos Muñoz, Hugo E. Silva
https://doi.org/10.1016/j.regsciurbeco.2020.103616
We investigate the impact of fare-free public transport on travel behavior by randomly assigning a pass to workers in Santiago (Chile). The pass allowed them unlimited travel for two weeks, as opposed to paying the regular fare of approximately US$ 1 per trip. The main impact of fare-free public transport is an increase in overall travel of 12%.
Finally, we find that having access to a fare-free public transport system did not have an impact on car trips. In particular, we did not find a significant effect on total car trips in the two-week treatment period, on car trips made during peak periods, or on car trips to travel to work or return home. Therefore, we find no evidence that fare-free public transport decreases negative externalities or that it increases public transport crowding during peak periods.
So a 12% increase in trip generation when fares were set to zero. Therefore, a smaller fare discount less than 100% would likely lead to an even smaller trip generation. The 12% result is similar to the ~ 15% result seen in the QLD Governments 50c fares policy.
Off-peak Fridays trial made no difference - Transport for London
Tom Edwards
Transport correspondent, London • @BBCTomEdwards
No noticeable difference in patronage when fares were free before 9.00 am on Fridays, and off-peak discount applied to the whole Friday.
Notes
Transport for London (2024), Off-peak Friday fares, Our trial of off-peak fares on London’s transport network between 8 March and 31 May 2024
offpeakfridaystrialreportacc.pdf (4.9 MB)
Queensland 50c fares and fines
More in depth look at free or near free PT and why it isn’t a best-in-class solution
Before 50c fares, Brisbane (and other Australian cities to a lesser extent) was highly criticised for having one of the most expensive PT systems in the world where passengers would pay up to 60$ a week. Driving was a very attractive mode of transport for many people purely because it was cheaper than PT.
50c fares were implemented to address Brisbane’s overpriced PT, and PT has become more attractive to people with decent access to PT (although not the majority of Brisbane residents). The trains and BUZ buses seem to have quite a bit of passengers in them even during off-peak where frequencies are 30min (only my observation when I catch PT myself). So the next step is to increase the off-peak frequencies to every 15min post CRR to see a more significant growth in patronage.
1 Like
Well, it’s only overpriced relative to the poor service on offer. People are happy to pay more to get more, which is why 80% of persons making journeys pay ~ $40k for a car (possibly two) and ongoing running costs despite buses being 50c. Discounts to basic fares can cover anything else.
The large draw of $300 million p.a. comes out of the transport budget, which means it’s not being spent on things we actually need, like frequency or service extension.
Free Fares Policies: Impact on Public Transport Mode Share and
Other Transport Policy Goals
Nils Fearnley
Institute of Transport Economics, Norway
[email protected]
International Journal of Transportation
Vol.1, No.1 (2013), pp.75-90
Abstract
This article investigates the merits of free public transport as a means to achieve a number of transport policy objectives, including mode shift towards public transport. It outlines some political and societal motivations behind proposals for free and low fare schemes, and presents key economic principles for public transport pricing.
Examples of free fare schemes mainly from Europe are summarised and their impacts synthesised. Although free public transport at a first glance may seem attractive both from economic, social and environmental perspectives, the message learnt from a number of schemes is that free public transport offers poor goal achievement in all these respects, and at a high cost. The main effect is a huge growth in patronage, up to 13-fold increase is reported, of which the larger brunt is shifted from walk/cycle, or induced. The effects on car traffic levels are marginal and typically they are offset already after a few years’ traffic growth.
Successful free public transport schemes are those whose goal is mainly to grow patronage. Congestion relief, social and environmental benefits are best achieved with more targeted measures, or in combination with such measures.
(bolding added)
Research from Norway looking at mostly European examples suggest minimal impact on traffic, and most new trips coming from reduced walking and cycling. This isn’t the substitution we’re aiming for.
Also does little to take traffic off the road, which just returns through induced demand.
In addition to trip generation there is also considerable mode substitution. The low cross price elasticity of car use with respect to public transport fares means, however and according to Storchmann [13], that free public transport is unsuitable instrument for reducing car use and its external costs. Motorists’ behaviour and mode choice depend very little on public transport fares
(bolding added)
With respect to mode shift, the general picture observed in most cities is that the source of the increased passenger numbers is overwhelmingly people who alternatively would have walked, cycled or not travelled at all. There is also a large group that alternatively would have used other public transport services. Very few come from car.
Notes
Research into a sudden price drop on PT in Spain during September 1 and December 31, 2022 also shows little to no environmental benefit as well (as measured by air quality indicators).
Free rides to cleaner air? Examining the impact of massive public transport fare discounts on air quality
Daniel Albalate, Mattia Borsati, Albert Gragera
Economics of Transportation, Volume 40, December 2024, 100380
Abstract
We quantify the effect of public transportation fare subsidies on air quality by exploiting the sharp discontinuity in the cost of ridership introduced by policy intervention. We identify this effect by taking advantage of four months of massive discounts for transit services introduced in Spain on September 1, 2022, as part of the national plan to tackle the global energy crisis.
Across pollutants and specifications, we find no evidence that low-cost or free-of-charge public transportation financing schemes have improved air quality. Our results reveal that measures aimed at reducing transit prices may fail to achieve the claimed environmental benefits through a modal shift from private to collective modes of transport, which suggests that massive fare discounts may not represent an efficient allocation of public funds.
(bolding added)
Notes
Even if the services have been upgraded to 15min frequencies of all train lines, the level of service will still not justify the expensive fares Brisbane saw before 50c fares. A level of service that justifies the previous fare system would be something similar to the like the world class PT systems of Tokyo, Singapore, Taipei, etc.
That level of service is simply unachievable with Brisbane’s current PT infrastructure. To build such PT infrastructure that can accommodate world class service would cost billions, and the 300mil$ commuters saved would not be enough to build it.
I feel like main reason people choose to drive over taking PT is that it is the most direct path from where they are to where they want to go. For example, if someone who lives in Deagon works in Wynnum, they could drive on the Gateway and Brisbane Port Motorways that lead almost directly from Deagon to Wynnum. By taking PT, passengers would have to travel all the way into the city, and all they way back out to Wynnum. Building PT infrasturcture that can accommodate journeys like this around Brisbane will cost a bit more than 300mil$. For people travelling to the CBD, PT is the most attractive option because PT is the most direct way to the city that rarely gets stuck in congestion. 80% of people travel to the city by PT as a result.
Edit: Furthermore, I definitely do not want to go back to spending more on PT with the previous fares like the majority of Brisbane PT users. And 50c fares is a very much welcomed by the PT users as well.
1 Like
Well an alternative is that we could set fares similar to what Perth has. They do have 15 min trains everywhere, and a 50c fares policy was not required to achieve it. And their train and per-capita PT use is higher than ours.
That level of service is simply unachievable with Brisbane’s current PT infrastructure. To build such PT infrastructure that can accommodate world class service would cost billions, and the 300mil$ commuters saved would not be enough to build it.
Sure, if one takes an infrastructure-based view that could be true, however the main issue in SE QLD is mainly lack of service during off-peak on existing infrastructure.
This generally only requires funding more service in the off-peak (including more HF or BUZ routes), something that is entirely possible with $300 million p.a. (some infra constraints are acknowledged with specific train lines such as Beenleigh and Cleveland). Based on $6 million p.a. to fund say a BUZ route or top up, that would be about ~ 50 BUZ services.
In general, a 50c fares policy isn’t necessary to get a service boost. Which can just be funded directly.
And 50c fares is a very much welcomed by the PT users as well.
But its not welcomed by motorists, which is what matters if we would like to shift people out of cars and into PT. See, 50c fares on PT means very little to them. This is why spending $300m p.a. on improved frequency and service extension into underserviced areas should have come first before considering this policy.
300mil$ would improve frequency to how much? Every 15min? Even if that lost ticket revenue has been used to upgrade PT frequencies to every 15min, it still would not be enough to justify passengers spending up to 60$ a week. Furthermore, since the QLD government already pours billions into road projects with no gain, they could divert that money onto improving service frequency and extending high-frequency services instead of revoking a popular PT fare policy that benefits the PT users of SEQ.
2 Likes
Well, suppose Perth decided to cut all day 15 min bus and train frequency in half. And in some Perth suburbs, bus service would be wound back to being hourly, or not provided at all.
This would create a city with an end state more like the current Brisbane scenario.
With the funds saved, a flat 50c fare would then be offered to Perth PT users as compensation.
Would this be a good policy? Its the same principles, just in reverse sequence.
Everything has a trade off and an opportunity cost.
This isn’t a zero sum game @Metro. You can have cheap and affordable fares and high frequency services. It’s just a matter of the Governments priorities.
Affordable fares benefit those who already have access to frequent PT.
Higher quality services benefits those who don’t yet have access to frequent PT.
Crucially though, affordable fares will also benefit those who gain access to frequent PT in the future. If we are able to keep affordable fares while also increasing service quality, you will undoubtedly benefit people more and make PT more attractive than if you just had one or the other.
My 2c is that 50c fares are a good incentive for those who have access to high quality PT, and that we shouldn’t return to the previous fare structure we had. That doesn’t mean I’m opposed to modest increases in the medium to long term, but it must remain affordable and reasonable.
7 Likes
Suppose that the QLD Government cut the funding for road projects like the Coomera Connector, Bruce and Gateway Motorway widening, Bruce Highway Western Alternative, and Gympie Road Bypass Tunnel, and they would use the funds to build much needed PT projects and service improvements. The money that could be used to fund service improvements may be allocated to something else that doesn’t really provide much benefit for the society of QLD as a whole.
2 Likes
The research is interesting to note. I’d suggest the variations in different contexts across cities and regions and the quality of public transport networks would make generalising them hard.
Regardless, dealing with the reality we have in QLD now - fares are 50c and I can’t imagine arguing for higher fares ever being productive to achieving BTQ’s goals and we should absolutely not start advocacy from a position of accepting there can be no further investment in PT without users having to pay more.
$300 million is nearly loose change when it comes to road projects.
3 Likes
Assessing The Affordability Argument
The affordability argument is often put because, well, who would want to argue that fares should be unaffordable? Nobody, of course. On reflection, assessing a policy requires answering questions on how to assess affordability both in absolute or relative terms.
It also creates a question around why the affordability measure should be administered through public transport fares specifically, rather than by just providing cash directly to Queenslanders that could be spent on anything (e.g. groceries, housing costs, or electricity), or some other broad measure (e.g. increased electricity or water rebates).
For example, during the 2007-2008 Global Financial Crisis, the Australian Government (Rudd Administration) deposited cash directly into citizens bank accounts as a relief measure, and it was able to be spent on anything. Not just bus or train fares.
From early April, anyone earning $80,000 or less will be paid the full $950; people earning between $80,000 and $90,000 will get $650; and those earning between $90,000 and $100,000 will receive $300. (see SMH newspaper reference in notes).
Fares may be high or low internationally, but earned wages also vary quite a bit across countries too. A country might have high fares, but also have high incomes, so its a good idea to take this into account. In some countries, such as Hong Kong, fares are low because they are cross-subsidised using TOD development revenue, clearly an option not applicable in the Brisbane case.
(bolding added)
Let’s look at this affordability claim a bit more closely.
The 50c fare applies to both those who need it and those who want it
Even if fares were higher than desired, that would only justify a temporary fare reduction and/or expansion of the eligibility criteria for discounted fares. It doesn’t itself establish a case for dropping fares permanently to the extreme level of 50c.
And it also doesn’t explain why the measure is untargeted. The previous fare structure did make this distinction - discounted fares applied to those who needed it, and full fares applied for everyone else.
I would invite those who support the 50c fare policy to take the opportunity to explain what the justification is for the policy being a) permanent and b) continuing to be untargeted.
How can we measure and thus assess affordability?
The Fair Work Commission sets the minimum wage as follows. Casuals receive a further 25% loading on top of this.
As of 1 July 2025, the National Minimum Wage is $24.95 per hour or $948 per week.
An affordability measure can be constructed by estimating how many minutes of work is required at the lowest wage to pay for the return trip PT fare.
$60 per week over 5 days → $12 per day
($12 daily return fare / $24.95 hour of work) = 0.48
60 minutes x 0.48 = 28.8 minutes → round up to 30 minutes
From this measure, a worker working a single hour at the minimum wage is able to cover the return fare cost in the first 30 minutes of their shift.
Notes
https://www.fairwork.gov.au/pay-and-wages/minimum-wages
Just to assure you, the BTQ Policy Platform does not propose to increase fares 
That said, it is not difficult to see how increased operational costs could act as a disincentive for the Queensland Government to expand service.
Supply and Demand: In general, when the cost of something rises, less of that something is supplied.
If the cost of producing the same PT service increases then… ?
And as we have seen with infrastructure costs, when these rose for the Sunshine Coast rail line or the Gold Coast Light Rail, the Queensland Government responded by simply cutting them back. I guess one would have to ask then, why would it be any different for operations and operational costs then?
Some may say that increased patronage could indirectly stimulate more service, and I hear that. But as the patronage level is essentially more or less the same as pre-covid, the added patronage from the 50c fares policy is simply not high enough to trigger further additional service by itself.
How are they supposed to define who wants 50c fares to users who need 50c fares? Isn’t the PT experience supposed to be balanced for all users, whether they’re frequent users or not? Why would they have a select group of people enjoy 50c fares while other users would have to spend up to 60$ a week? It’s just not fair.
I believe the cancellation/enshitification of the GCLR4 and DSCRL is primarily for political reasons, rather than for construction and operational costs. Both projects were backed by the ALP and, had that remained in power, would’ve progressed further to start construction of GCLR4 and have DSCRL bee all rail to Maroochydore.
The cancellation of GCLR4 is done be the LNP in an attempt to call out the ALP for their ‘sham consultation’ and that they probably never supported the project in the first place.
The enshitification of DSCRL is the LNP trying to prove that they can deliver PT to Maroochydore in one fell swoop before the Olympics in an attempt to call the ALP out for only delivering stage 1 for the Olympics.
Both of the project would be multi-million dollar projects, and 300mil$ would only be a very small fraction of the budget required to build the new infrastructure.
Considering that the state government pours billions into roadways for no gain for the road users, they would use that money to build funds to construct much needed PT projects and to improve PT services without the need to revoke 50c fares.
2 Likes