Myth: Park-and-Ride will encourage public transport use

Myth: Park-and-Ride facilities will encourage public transport use
Fact: Expecting people to drive to railway stations puts an artificial limit on the number of passengers the train system can support. Car parks are expensive to build and maintain, are space-intensive, lock up valuable real estate near stations, and require public transport users to own just as many cars as if public transport didn’t exist. In well-patronised rail systems around the world, most train passengers arrive at the station by feeder bus or tram, not by car; it would simply be impossible to provide enough car parking to get all these passengers there by car instead.

One of the most visible aspects of most Melbourne railway stations is the acres of car parking that surround them. Our public transport authorities are constantly expanding these car parks, thinking this is an easy way to encourage more people to catch trains; certainly easier than feeder buses, which require planning and funding. And who could disagree? The car parks are well used, and anyone who catches a train from the suburbs in the morning, or gets off a train in the afternoon, can see for themselves that a lot of passengers drive themselves to and from the station.

But looks can also deceive. Travel to work by train in Melbourne increased by around 50% between 2001 and 2011, and at the time of the 2011 Census stood at around 184,000 journeys each working day. The following table shows how these commuters got to the station on Census day in 2001, 2006 and 2011.

2001 2006 2011
Access mode No. % No. % No. %
Walk 68,682 58.0 87,216 61.3 112,099 60.8
Feeder bus or tram 22,201 18.7 26,718 18.8 42,022 22.8
Car driver 19,274 16.3 20,208 14.2 22,393 12.1
Car passenger 6,493 5.5 6,041 4.2 5,553 3.0
Bicycle 1,110 0.9 1,282 0.9 1,674 0.9
Other (mainly taxi) 733 0.6 888 0.6 730 0.4
Total 118,493 100.0 142,359 100.0 184,471 100.0

(Source: Australian Bureau of Statistics, Method of travel to work (full classification list) – 2001 and 2006 census, TableBuilder – 2011 census. Totals may not sum exactly due to rounding. Classification of access modes is according to the methodology of Mees and Groenhart, Transport Policy at the Crossroads: Travel to work in Australian capital cities 1976-2011, RMIT University, December 2012.)

As the figures show, barely one in six Melbourne rail commuters gets to the station by car – down from one in five in 2001. On the other hand, a clear majority of train travellers walk to the station, while an increasingly larger proportion arrive by bus or tram compared to those arriving by car (despite the generally very poor provision of these services).

What is also particularly interesting is that the vast majority of new train passengers in 2006 and 2011 (those who weren’t using the train in 2001) also either walk to the station or use feeder transport. In fact, the number of people walking to trains increased by 63% between 2001 and 2011, and the number using feeder buses or trams by an impressive 89%. Meanwhile, the number driving or being driven to the station increased by just 8%.

In absolute numbers, the recent increase in train patronage is dominated by those living within walking distance of a station. Feeder services, poor as they are, have also played a significant part in delivering more people to trains, especially since 2006 when a number of frequent ‘SmartBus’ services were introduced. On the other hand, the number of park-and-ride and ‘kiss-and-ride’ passengers has been relatively static in absolute terms and declining in relative terms, despite the government’s multi-million dollar efforts in building more station car parks.

This hasn’t stopped the road lobby using park-and-ride as a way of having their cake and eating it too: calling for more roads and more car parks as a way to get people to railway stations. This has led to some curiously ironic, or outright misguided, statements from community leaders who have been bamboozled into thinking that the way to improve public transport is to build bigger roads to drive cars on.

The State Government has announced $18 million in funding for the [Plenty Road duplication] due to start in March next year that will provide double lanes to Gordon Rd….on top of $14 million to duplicate Plenty Rd from Centenary Drive to Bethany Court….

[City of Whittlesea] Mayor Sam Alessi said the upgrade would result in more people using public transport.

We want to support the target to have more people using public transport by 2010….and the extension is needed to get people to where the transport is, he said.

—“Road Hope in Sight”, Whittlesea Leader, 27 July 2005

The cost of petrol has increased the demand for public transport. We must improve parking.

—Cr Brian Oates (City of Casey), Cranbourne Leader, 5 October 2005

If we are going to have better public transport, we need better infrastructure and better roads are certainly a big part of that.

—Federal Opposition Leader Tony Abbott, Four Corners, 16 August 2010

How do people get to railway stations in cities with ‘world’s best practice’ public transport? Comparative figures show some striking differences between Melbourne and other cities. Paul Mees, in A Very Public Solution, compared access to stations by Melbourne’s 88 million train passengers in 1991 with Toronto’s 302 million train passengers:

Melbourne Toronto
Access mode Passengers Percent Passengers Percent
Car 17,000,000 20 9,000,000 3
Feeder bus or tram 9,000,000 10 230,000,000 76
Walk or cycle 61,000,000 69 60,000,000 20
Other (mainly taxi) 1,000,000 1 3,000,000 1
Total 88,000,000 100 302,000,000 100

(Source: Mees, A Very Public Solution, p.232)

These figures are for all passengers, not just those travelling to work. They show that for non-work trips the share of people walking to the station is even greater, and the share of those getting to the station by car, bus or tram even less. (This is to be expected, since station car parks are quickly filled in the morning with the cars of nine-to-five commuters, while feeder services in Melbourne are even poorer off-peak than in peak hour.)

The actual number of people walking or cycling to stations is the same in both cities, though in Toronto this number corresponds to only 20 per cent of passengers, compared with Melbourne’s 68 per cent. Fewer people get to the station by car, largely because Toronto has fewer parking spaces – 11,000 compared with 27,500 in Melbourne in 1991. However, in Toronto those parking spaces are concentrated at fewer railway stations, which means that station car parks in Toronto take the form of huge multi-storey edifices, which fill up daily and yet only cater to 3 per cent of train travellers.

The most striking difference between Melbourne and Toronto is the number of people arriving at the station by feeder bus or tram. Astoundingly, over 25 times as many people got to the station this way in Toronto as in Melbourne. This disparity helps explain the success of Toronto’s public transport system, which in 1991 carried one-quarter of all trips in Metropolitan Toronto, and one-third of all work trips. As in Melbourne, around 80 per cent of Torontonians don’t live within walking distance of a railway station. But in Toronto this 80 per cent slice of the population has a fast, frequent feeder bus network to access the rail system. In Melbourne a small portion of this 80 per cent has some semblance of this, thanks to bus network improvements in the past decade. But the rest – the vast majority – has no such service and so does not use public transport much at all, except perhaps in peak hour to commute to work.

This key difference in Melbourne between the transport habits of people who have a station within walking distance, and those who don’t, also explains why the proportion of people driving to the station is usually thought much greater than it really is. If you’re a typical Melburnian, you probably don’t live within walking distance of a station yourself, and chances are most of your friends don’t either. If any of them do use trains, they’ll more likely than not drive to the station. You’re less likely to know train users who walk to the station, simply because these users, while more numerous, live in a different part of town.

Given the geographic similarities between Melbourne and Toronto, the latter’s 1991 ‘benchmark’ of 25% of motorised trips by public transport is a realistic goal for Melbourne three decades later. But to remotely approach this, a way needs to be found to get people to railway stations who live beyond walking distance. There are basically three methods: car, bicycle or feeder bus.

Driving to the Station: a failed strategy

For decades the Government has relied unsuccessfully on the ‘car’ option. Not only does the evidence show this is an ineffective strategy: there are a number of obvious problems with trying to accommodate a large population of public transport users by getting all (or even most) to drive to the station. While it’s fine to provide some park-and-ride capacity, relying on it exclusively is likely to fail for the following reasons.

It’s never going to be enough. This is a familiar problem: the car is a perfect method of transport as long as you’re the only one using it. Currently, around 6 per cent of Melburnians are regular train users, and 15 per cent drive to the station; Melbourne’s railway station car parks are quite up to housing the cars of this 1 per cent of the population (15 per cent of 6 per cent). But what if not 1 per cent, but 25 per cent of Melbourne’s population wanted to park their cars at railway stations? There wouldn’t be enough land or enough concrete to handle the task!

The daily battle for a park is causing considerable frustration among commuters, and at Brighton station, 240 spaces were already filled by 7.30am yesterday…. At Syndal station, in Glen Waverley, the 340-space car park is packed by 8am. Parking in nearby streets is limited, so commuters park further away and gallop down the hill to catch their train.

Rob Varney has used this line for 40 years and said increased patronage left few spaces. He used to get the train from Glen Waverley station, but changed to Syndal when he could no longer find a park at 7.10am. Now he is forced to park further and further from the station.

—“Latest problem on train network – car parks”, The Age, 27 May 2008

Overcrowding at Watsonia railway station carpark continues to be an issue as politicians argue over where the cars come from. Banyule Mayor Tom Melican and Bundoora state Labor MP Colin Brooks have different ideas about why the carpark is packed by 7am each work day…. Colin Brooks’s office conducted an informal survey of more than 400 commuters over two mornings last September, which showed local residents were the highest users.

—“Divide over Watsonia station parking woes”, Diamond Valley Leader, 10 April 2009

Star reported on 16 October 2007, just two weeks after the electrification of the Craigieburn line, that residents and Hume councillors were seeing shortfalls in the limited 300-space car park at the station. Then-councillor Moya White said at the time that the car parks were full by around 7.30am on weekdays and that she had received a number of calls from residents about the parking shortfall.

—“Safety risk at station”, Hume Star, 14 April 2009

More than 719 travellers have this year been fined for parking in no-standing areas around suburban stations. A further 49 people were given infringement notices for parking on the footpath at train stations. Another 26 commuters were fined for illegally parking in a permit zone.

—“Car space shortage driving train commuters to park illegally”, Herald Sun, 2 December 2011

Car parking at stations was oversubscribed back in 1969, when the Metropolitan Transportation Plan called for a massive expansion of car parks, and it is still oversubscribed today despite a fivefold increase. A 2009 study by the Department of Transport’s Paul Hamer found that demand for car parking at stations exceeded supply by around 50%, and continued to exceed supply even after more spaces were provided to ‘meet the demand’. Like building roads to relieve congestion, the provision of parking creates its own demand, and so increases the problem instead of reducing it.

Vehicle counts [at stations] in 2007 roughly matched the total number of parking spaces that were provided once the [park and ride] upgrades were completed. However, in the year between observations, parking demand increased further meaning that the car parks at each station were again heavily over-subscribed.

—Paul Hamer, Road and Transport Research, March 2010

There’s also the problem we mentioned above, that station car parks fill up with the cars of nine-to-five commuters leaving no spaces available for those who need to travel during the day. Good public transport is focussed on all travel, not just peak-hour commuting, but park-and-ride is a particularly ineffective way to cater for weekday off-peak travel.

And even if we doubled the size of every station car park in Melbourne, the very best we could do on the figures above is increase train patronage by some 16 per cent – from 6 per cent of the population to 7 per cent.

Such a modest increase seems hardly worth the very substantial cost of providing all that extra parking. Which brings us to possibly the biggest practical problem with depending on rail access by car.

It’s much more expensive to provide than often realised. As transport planner Vukan Vuchic has noted, car parking is one of the most heavily subsidised elements of the entire transport system. Multi-storey parking structures plainly cost money, but car parking can be unexpectedly costly even when such structures aren’t involved.

To take one example: in 2005, the government spent $2 million on an additional 120 car spaces at Huntingdale station. Depending how one views it, this amounts to either a $17,000 gift to each of 120 lucky car owners, or an attempt to grow patronage at a cost of $17,000 per additional train passenger.

Other park-and-ride facilities built or planned in the last decade have ranged in cost from $12,000 per space to a massive $68,000 per space – the latter for a multi-storey car park at Glen Waverley involving a $24.5 million outlay to replace 140 existing spaces with a 500-space structure.

In a much-publicised 2015 example, the government spent $10.8 million on a multi-deck car park to provide an additional 250 parking spaces at Syndal station. This example is particularly notable because it coincided with an ‘unsolicited proposal’ by private operator Metro Trains to build similar car parks at multiple stations, reportedly to help boost train patronage.

The prefabricated structures Metro were proposing were claimed to cost slightly less than what was built at Syndal – about $30,000 per space rather than Syndal’s $43,000 – but there is really no rational debate to be had when contemplating costs of this magnitude, which would have made an excellent case study for Donald Shoup’s classic text The High Cost of Free Parking.

Consider the hard numbers: even if each individual parking space succeeded in attracting a new passenger paying full Myki Money fares on 360 days each year (250 of them regular weekdays), the cash flow of $2610 per annum in 2016 dollars (over a 20 year lifetime at the 7% discount rate recommended by Infrastructure Australia) comes to less than $30,000 in ‘present value’ terms even with zero ongoing maintenance costs. Not only does this fail to recoup the outlay on the car park – on these unrealistically favourable assumptions – it provides no revenue to run actual public transport services!

There is of course one way such an outlay could make sense, and that’s if carpark users were charged a fee set high enough to at least recover costs. That would however appear to defeat the purpose at Syndal, which was to reduce pressure on the (free) parking in the street. Nor is it likely to win political acceptance at other locations.

Meanwhile, for a fraction of the $10.8 million outlay, one could buy and operate a small fleet of buses which would deliver local passengers to trains all day, not just in peak hour when parking spaces might be available. After all, 250 spaces represents less than 10 per cent of the 2,600 people who were estimated to use Syndal station each weekday in 2015, and anyone trying to drive to the station outside peak hour would still confront a full car park.

It can’t be relied on to increase patronage. Relying on park-and-ride to grow patronage assumes that every new driver using the car park is someone who used to drive alone to their destination – as distinct from someone who previously caught the bus, or perhaps carpooled with someone else. This was the fallacy behind the ‘successful’ park-and-ride station originally installed at Doncaster in 2002, which allows people to park at the entrance to the Eastern Freeway and catch a bus the rest of the way into the city. A review of the facility by the Department of Infrastructure in 2004 found that of those using it over the survey period, precisely one person was a new user of public transport. Every other person had previously caught a bus all the way from home to the city.

The real reason for the car park’s popularity was that it was located on a fare boundary. (This was in the days when National Bus Company had its own single-trip fare system incompatible with the Metcards used elsewhere.) This made it advantageous for a lot of former bus users to drive as far as the freeway, park for free, and then change to the bus and pay a cheaper fare. It certainly was a ‘success’, if the objective was to get people to drive more!

More recently, the above-mentioned 2009 study by Paul Hamer surveyed new users of park-and-ride facilities installed at Melbourne railway stations in 2008. It turned out that only one-third had previously driven all the way to their destination. And when asked to give a reason for changing their travel behaviour, most cited extraneous factors like moving house or changing jobs. For only a handful of people was the new car park decisive in attracting them to public transport.

A 2013 study in the Journal of Transport Geography (summarised in The Atlantic) confirms that in some cases park-and-ride can encourage car use at the expense of public transport and other modes. In Rotterdam, for example, only a quarter of park-and-ride users would do the entire journey by car if the facility were not available. The remainder would mostly revert to using public transport for the whole trip (as above), or change their travel habits, such as by combining multiple trips into one. A few would switch to cycling.

It’s contrary to good land use planning. There are compelling urban planning arguments against the wholesale expansion of park-and-ride facilities. Many railway stations in Melbourne are located in district centres, with a concentration of nearby commercial activity. We are fortunate in Melbourne to have such a pattern of rail-centred urban development. But this also puts a premium on real estate near stations, which means that expansion of car parks is not only expensive but also displaces other, more useful commercial activities. The cars of train passengers also add to traffic volumes in district centres, detracting from urban amenity.

It places pressure on car parking for other purposes. Because there will never be enough car spaces to accommodate all train users, giving people no alternative to driving to the station means that pressure is inevitably placed on parking for other purposes as well. Even with our current low patronage levels and higher provision of car spaces compared with Toronto, this is already apparent:

The traders around Main Street, Pakenham, and Wood Street in Beaconsfield are particularly concerned that [commuters] are using car spots that are meant for their customers. This poses a major problem because they are generally occupying these car parks for up to eight hours a day while they are at work.

—Cardinia Councillor Brett Owen, Pakenham Gazette, 24 May 2006

At Epping station they are all parking at Bunnings shopping centre. At Lalor they are all parking on the grass because there aren’t enough spaces there and at Thomastown they are all parking up side streets.

—Whittlesea Councillor Rex Griffin, Whittlesea Leader, 31 October 2006

We just think we’re going back to the bad old days – it’s a free car park. A lot of the people are coming from the outer suburbs. They get off at Bulla Road so they don’t pay the [CityLink] toll and they park as far in as possible, so they pay only zone one [train] fares. Everyone wants people to use public transport, but why should they park in our streets and ruin our quality of life?

—Merrill Shulkes, Moonee Valley Community News, 7 November 2006

Traders are getting angry and losing business because commuters are parking all day in spaces designated for shoppers. Parking in Mooroolbark is at a premium.

—Yarra Ranges Coucillor Terry Avery, Lilydale & Yarra Valley Leader, 24 March 2008

This is a huge problem; Croydon in particular is a disaster. We recently provided a large number of all-day parks for local traders, but they’re constantly being used by train passengers.

—Maroondah Mayor Tony Dib, The Age, 27 May 2008

The converse can of course also occur. Where there is already high demand for parking, park-and-ride facilities that are not ‘gated’ (or otherwise reserved for exclusive use by passengers) will inevitably be used for local shopping and other unintended purposes – sometimes called ‘park-and-walk’. Though this works to the (slight) advantage of local residents and traders, it clearly undermines the ability of park-and-ride to boost public transport patronage.

It entrenches environmental and social equity problems. Park-and-ride requires even regular public transport users to own cars, and has nothing to offer those who (for whatever reason) do not own a car. If the car is left at the park-and-ride station while the owner is at work, the car remains unavailable to other household members, so that a second or third car is still necessary. Meanwhile, the car itself is vulnerable to theft – so much so that ironically, police in Epping and
Dandenong are now encouraging commuters to certify that their cars are not used during the day, so that a car on the road during working hours can be identified as stolen! Is it really a good thing to require families to shoulder the cost of a car which must then sit idle for eight hours a day?

Park-and-ride also leads to an increased number of short car trips, which contribute disproportionately to air pollution owing to the ‘cold start’ effect. ‘Kiss-and-ride’, a variation in which a family member drops off the traveller and returns home, is no solution as it leads to increased empty running of private cars.

It undermines public transport use. Perhaps one of the biggest problems with the idea of driving to public transport is psychological. Once you’ve started your journey in a car, you’ve got little reason not to drive it all the way to your destination; a fact not lost on public transport users themselves.

On more than one occasion I have dropped the kids off at school but been unable to park on either side of the rail line near Mentone. I have driven to Cheltenham to find the same thing. Then, rather than waste the rest of the day looking for another station, the solution is to just keep going — into the CBD.

—Joanne Hughes (Mentone), Herald Sun, 19 May 2006

Had a meeting in the city last week, so I drove the car to Sandringham station about 10am. Drove around for five minutes looking for parking. No spots. Even the two-hour parking was full. I needed a three-hour parking spot so I could get into the city, have the meeting, come back. So I drove in. Cost of petrol: about $10 (same as a train ticket). Carbon emissions: hate to think (I have a family car).

—Bede Doherty (Cheltenham), The Age, 28 April 2008

And when it comes to encouraging car use while claiming to do the opposite – the goal of road-lobby spin doctors everywhere – park-and-ride (now separated from railway station access) is eminently suited to the task. In both Adelaide and Perth, planners have provided large car parks on the CBD fringe and encouraged their use through the provision of free bus or tram services. The planners’ spin draws attention to the well-patronised public transport service, neglecting to mention that this represents only the last mile of what are predominantly car journeys. If the car parks were located right in the CBD instead and the shuttle service dispensed with, overall traffic congestion and emissions would increase only marginally.

This was essentially the conclusion reached in a 2012 article in the journal Road & Transport Research. A team of five researchers found that after the South Australian Government opened a park-and-ride car park at the Adelaide Entertainment Centre, served by an extended tram route, car travel into central Adelaide actually increased. In fact nearly two-thirds of people using the new car park had previously travelled by public transport, but had replaced their public transport journey from home with a car trip instead. And because the tram from the car park to the CBD is free, the government actually lost all the revenue previously contributed by the public transport users.

Cycling to the Station: a healthy option

One alternative to park-and-ride is pedal-and-ride, where people use bicycles to get to the train station. This is an environmentally desirable option which is taken up by choice in some European countries such as the Netherlands and Denmark that have adopted an aggressive approach to traffic calming, limiting vehicle speeds in urban areas and restricting road space to create an urban environment conducive to cycling. Adapting this approach to Melbourne carries significant benefits and comparatively little cost.

Admittedly, it will be a daunting challenge to recreate the European experience in the outer suburbs of Melbourne, whose urban environments have been steadily moving in the opposite direction over decades. While pedal-and-ride is a mode choice compatible with the provision of feeder buses, it is incompatible with park-and-ride, as the latter cannot help but increase the volume of car traffic on roads leading to railway stations, and high traffic volumes and speeds are a major deterrent to cycling in Melbourne.

Provision of a comprehensive network of bicycle lanes and paths, and bike parking at stations – but not additional car parking – will help make cycling an attractive way to get to the station.

Bus to the Station: a public transport solution

The remaining alternative is feeder bus services. Part of what makes this so attractive compared with park-and-ride is that feeder buses have useful functions other than conveying train passengers to railway stations. Half of all our journeys are local trips, with the origin and destination within the same suburb, and when such journeys are made by car they also contribute disproportionately to pollution and congestion. A comprehensive bus network is essential to making public transport attractive for these local trips, as our Squaresville thought experiment shows. But as suburban activity centres and railway stations are often found in the same locations, the same buses that are useful for local travel are also useful as feeders to stations.

The feeder bus option is currently at a disadvantage relative to the car option, for reasons that are not at all hard to find (unless you’re a Victorian Government transport planner):

Jim Adamopoulos doesn’t catch the bus. If he did go by bus to the railway station at Albion, he would have to wait 20 minutes for a train because the public transport timetables are so far out of synch. Instead, every morning he drives his car from home at Caroline Springs to Albion.

Platform for change, The Age, 29 January 2008

This kind of non-coordination, together with low frequencies, limited operating hours, slow convoluted routes and an absence of bus priority in traffic, ensure that Melbourne’s buses are no more than a last-resort option, even for a relatively short trip to the nearest railway station. Improvements in frequency, coordination and priority up to a standard that is commonplace in European and Canadian cities could make buses competitive with cars, and superior to park-and-ride both in cost and convenience.

Park and Ride: an option, not a panacea

The problems with park-and-ride become apparent when planners rely on it as the only method of delivering passengers to public transport. This is not to say that it can’t be a useful backup option when applied in a limited number of locations to supplement a good feeder bus network.

While experience in other cities with high public transport use points to the importance of feeder buses, it remains true that some train users do prefer to drive to the station. Park-and-ride remains as a minority travel mode in Toronto, Vancouver and other cities with well-used public transport.

Many of the disadvantages of park-and-ride pointed out above can be avoided if park-and-ride facilities are located at those railway stations that do not have intense development nearby, and if park-and-ride is always viewed as complementary to feeder services and never as an alternative that makes feeder services unnecessary. This is how park-and-ride is used in Vancouver, to expand the catchment for rail services in fringe areas. And where feeder services within urban areas are inadequate, it is only good sense that their improvement take priority over the expansion of park-and-ride facilities for funding.


Return to index


Last modified: 3 January 2016