There is a lot of confusion out there about Park-and-Ride. Is it necessary for ridership? Are motorists entitled to it? Can it last forever?
Let’s start with the basic math.
- Really great transit generates high land value around stations.
- Free parking presumes low land value around stations.
It’s a contradiction.
When a transit agency provides free or underpriced parking at a station where the land value signals that there is a higher use, it is subsidizing motorists in two ways. First, it is forcing a low-value land use to prevail over a high value land use, and second it is making a much bigger investment in access by motorists than it makes in access by people who get to the station in other ways. I am using the word subsidy in exactly the same sense that any other artificial limit on price is a subsidy.
Obviously the problem is much worse where the rapid transit is of the highest utility and quality and where the ambient land value is therefore higher. This is why free Park-and-Ride is much harder to justify on high-frequency metro systems than on infrequent outer-suburban commuter rail in most metro areas.
Low-cost Park-and-Ride can make great sense where a station area is undevelopable (floodplains etc). There’s also an important role for distributed, small-scale Park-and-Ride created by sharing existing spaces. Transit agencies often make deals to share parking with land uses that peak on weekends and evenings, such as houses of worship and entertainment venues. These are great ways of providing some car access at very little cost to the public.
But the law of supply and demand generates some facts about free Park-and-Ride that many people don’t want to hear, but that we really can’t protect them from:
- Free parking at a high-utility rapid transit station is a price subsidy, exactly the way the Soviet Union’s caps on retail prices were. It has the same effect, which is to cause problems of supply: Empty shelves in Soviet grocery stores, Park-and-Rides that fill up at 7 AM. If a commodity is priced too low in a condition of high demand, its supply will be exhausted, making it unavailable.
- A Park-and-Ride that fills up at 7 AM is effectively one that doesn’t exist over much of the time period when it’s supposedly needed. This loss of utility for people who travel later is a direct consequence of the price subsidy, as the artificially low price prevents the transit agency and its customers from reaching a market equilibrium where supply and demand of parking are in balance. (This equilibrium, of course, would be optimal for both ridership and revenue. Parking that fills up too soon drives away riders as effectively as no parking would.)
- The claim that Park-and-Ride is needed to attract riders is true only in the earliest phases of development, or on transit services with limited utility like peak-only express service. Once land value rises in response to transit access, the highest source of ridership is also the economically highest use of the land: dense, transit-oriented development around the station combined with good provision for the space-efficient forms of access (i.e. everything but Park-and-Ride). This is why Park-and-Ride is often a logical interim use of land, but not one that you should plan on having forever. Once a city has grown in around a transit system, there may be little Park-and-Ride left at rail stations, and only massive, distorting subsidies will make it free.
- Preventing high-value dense development on naturally expensive station-area land forces that development to locate away from the rapid transit system instead, creating a less sustainable urban structure in which more people and businesses lack excellent transit options.
- People who take buses or bikes or their feet or Kiss-and-Ride to a rail station are being mathematically correct (and fiscally conservative) when they object to free Park-and-Ride at high-demand stations, especially if the agency is not offering a corresponding subsidy to their own preferred modes of access, which all use scarce space more efficiently.
- All of these problems around Park-and-Ride can be resolved only by charging a fair market price for the rental of expensive, publicly owned real estate. Once parking is priced that way, it can remain the best use of valuable land. As always, the problem is not parking, per se. The problem is the market distortion arising from the subsidy.
It’s easy to feel entitled to a free Park-and-Ride space. But nobody can repeal the law of supply and demand, and that’s what we’re dealing with here.
Photo: Park and Ride at Auckland’s Constellation Station, Auckland Transport
Hi,
There’s no reason for you to have known this, but building car parks on floodplains is usually a bad idea. The rainwater-management services the land provides are usually far more valuable than the market price of the parking.
Yours,
Jonathan Hallam
Science, bitches. Or rather, math, bitches.
Beautiful smackdown, an instant link-reference classic. Bravo sir, and thank you.
@Hallam couldn’t you still have the drainage advantages by building the parking with a gravel macadam surface? Nobody should be driving fast in the lanes of a parking lot anyway.
That would last about a week in Seattle, and you’re still compromising the rain drainage. (Which doesn’t necessarily keep low suburbs in our Green River Valley from trying it. Safeway built a semi pervious parking lot in my neighborhood last year as an experiment for their LEED program, and it’s already being rebuilt.
Now that I think about it, providing insanely cheap parking at nearly every station is almost exactly what’s wrong with BART today. They recently opened a new parking garage at MacArthur station as part of a “transit oriented” development and many of the existing stations even in the inner East Bay are surrounded by parking, single family homes, and empty lots. It’s like land use barely changed since 1973!
@Brian. Possibly. Unfortunately I now can’t find the link to the document I was recalling, which was a study of the value of the environmental services provided by nature.
Very nice piece! Even though I am not an economist I like to think of the (wrong) way people are trying to solve shortage of parking with an economic analogy. When there is shortage of parking the main solution up to now has been “let’s build more parking”. However, when there is shortage of money (if I may put it that way) someone who will say “hey, let’s print more money” will be treated as crazy (or a money forger :))
That LA Times article is really irresponsible for not mentioning the price of parking at those LA MTA stations. The idea that the Red Line subway in LA should be primarily for park-and-ride customers is nuts.
Most older subway/rail systems don’t have the issue of Park ‘n Rides because those networks were built before the automobile became widely available and inexpensive. I’m from New York City (The Bronx to be exact) and other than a couple of “car parks” on the eastern edge of our subway system (in Queens and all built in recent years), there’s virtually no free parking for motorists to use.
One other thing on your point about a parking lot that fills up at 7am not “exist[ing] over much of the time period when it’s supposedly needed”: The people who take it before 7am are driving during hours when traffic is relatively less congested, which forces those who drive a little later, when there are more cars on the road, to drive even if they’d prefer not to. As far as opening space up on the roads, if your park and ride is filling up before peak hours then it’s contributing just about nothing.
Thanks for putting into words a more subtle version of what I declared it, which was “Motordom Marxism” (of which there is quite a bit in Adam Smith’s alleged utopia, the USA)
In LA, its not just park and ride that takes up too much space; its the stations themselves. On the subway lines in particular, stations consist of the portal(s) itself surrounded by a concrete area of some sort with no functionality whatsoever. They’re not parks and I woulnt even go so far as to call them plazas. There aren’t even any benches in most cases. Sometimes its not so bad- at wilshire/western, the station is circled by 3 or 4 shops and there are these lopsided sculpture things for people to sit on. But at the other end of the spectrum we have vermont/santa monica (a picture of the station aquired from google images tells you all you need to know). Metro kinda has an excuse- when the subways were built in the 90s TOD wasnt a thing and the land around the stations wasnt valuable. But TOD is a thing now and I expected station designs for the new Purple Line subway extention to reflect that, especially since the land it ventures into is, shall I say, considered more valuable than than where it goes now. However the renderings reveal the same waseful design. I dont see whats so hard about building a staircase into the ground next to a basic elevator on the sidewalk, like in new york and san fransisco.
Everything you say is true, but it ignores one important point. That if you wish to improve transit ridership in low density suburban neighbourhoods, you have exactly two likely-to-be-successful strategies to choose from. They are firstly, densification of the corridor (which effectively lowers barriers to people to living near the station), and secondly, park-and-ride (which effectively lowers barriers to people parking near the station).
I personally would prefer to see the densification solution prevail every time, but sometimes that isn’t desirable to the residents or governments of a particular area. If you write off park-and-ride out of hand, then you implicitly encourage the use of ineffective strategies to increase ridership in places where densification is considered inappropriate for whatever reason.
Many localities do bar or hinder development around high-quality, high-demand transit. It can happen in a variety of ways. Sprawling station designs with underpriced park-and-ride are common. Other barriers are: zoning that restricts uses; zoning that mandates low density; excessive parking requirements for commercial uses; wide, fast arterials occupying much of the station area; etc.
Localities that do those things pay the price in many ways. They get lower ridership; artificially depressed land values; less property and sales tax revenue; more sprawl, VMT, and air and water pollution; and so on. The large and growing market for TOD living is not served. Localities that provide high-quality transit service and also hinder development near stops and stations are earning a very poor return on their investment.
@Jarrett, A big thank you for this piece. I had been pondering how to write the reporter, but you have nailed it.
@Brian. Yes, the Macarthur BART “transit village” is a travesty. The parking garage is a huge waste of both money and land–well sound damped office space would be better. Worse yet, the 1 and 2 story buildings along Telegraph and Macarthur were not bull dozedso even more “transit village” isn’t happening.And all should note how nothing is being done on the west side of the station/freeway “tracks”.
@imustfeed – You are exactly right. Metro seems to have a love affair with wide open, un-shaded concrete plazas. Their renderings always show lots of people milling around “enjoying the sunshine”. Go to any transit or bus stop for half the year and spot the tight crowd huddled in the tiny shaded spot trying to avoid the sun.
I think the “plaza” design came from Metro acquiring whole lots for the station construction. But they seem oddly unthoughtful about what to do with all that space once construction is done. In not one station I’ve seen have they built even a simple kiosk type retail space that would use the space well, bring more people into the area, and generate revenue for the whole system. I really don’t understand what their aversion to it is other than that they view themselves “transpo professionals” not property managers.
What if the station co-located at a suburban shopping mall? There is a large sunk cost for parking, the demand for which peaks at weekends and evenings, so having cars parked there during a weekday seems like an efficient way of doing things. Better still, the shopping mall serves as a “transfer station” which can capture a lot of shopping traffic.
Weirdly enough, there is a shopping mall at San Bruno BART, but as I recall BART had its own parking structure and the Mall was at odds with anyone who might park at the mall to ride the train. For some time there was even a fence between the train station and mall entrances for some bizarre legal liability reason.
BART also has a station at Bayfair Mall in San Leandro–the two facilities are separated by a creek, and at El Cerrito Plaza where the station is separated from the shopping center by a city street. Efforts to repurpose unused parking space for housing was blocked by NIMBYs; so in neither case are riders encouraged to use the parking lots.
Great article
One minor correction – the photo is the park and ride at Albany, not Constellation. But yes it is on Auckland’s Northern Busway
Melbourne, Australia has a love affair with park+ride at stations, with governments constantly expanding carparks rather than provide frequent feeder buses, many running 30 – 40 min peak frequencies connecting with trains every 5 – 10 mins. Where good services are provided, people use them – most notably the 465 & 467 routes in the middle north around Essendon and Moonee Ponds.
This creates a vicious circle, with public transport limited in suburbs away from rail, creating a culture of driving – good feeder buses could help those making working in the suburbs too, catching a bus to the station to meet another feeder route in the counter-peak direction or catching a train between two routes.
Those parking before 6:30am only pay for a single journey (rather than return) thanks to the free early bird fares encouraging travel arriving the CBD before the AM peak – thus filling the carparks up before paying customers in the peak…
so how about both- feeder buses and parking?
Toronto has parking only around a few of its suburban transit stations and it is not free. It is almost impossible to get a spot after 9:00 a.m. at the stations I sometimes use. In the late ’90s the province changed the tax rules and the TTC made the parking “free” if you had a transit pass. The tax rules were changed, the TTC went back to paid parking and the lots were almost empty for a few months as people didn’t want to pay for what used to be free. The lots are once again full.
Most of the subway riders in Toronto take surface transit at one or both ends of there journey. This is possible because the TTC runs a grid system where over 99% of surface routes make a connection with the subway and have a free transfer (connection for Jarrett.) Rail is not the only form of transit around and if you do not provide good feeders to it then you will need “free parking.”
GO transit, the suburban commuter rail operator, provides free parking at almost all of its stations and is the largest parking provider in the Greater Toronto Area, over 60,000 spaces I believe. They are now building massive parking garages at an average cost of $42,000 Canadian per space. They do not have a co-ordinated feeder system with local transit providers so car is the way most people get to the station.
There are a few malls at major transit stations and they discourage people from parking and taking transit as their parking lots would be full before they opened.
All the LA metro “plazas” are TOD candidates. Wilshire / Vermont is obviously a successful one, and the W Hotel at Hollywood / Vine is another.
BART and GO are two transit agencies who are not in the business of operating local feeder bus services and thus would have to coordinate with local bus agencies, some of which they may have antagonistic relationships with. It is probably less of a hassle for them to build parking then to try to achieve some kind of coordination.
For Los Angeles Metro, coordination is not an issue but you have to wonder if it is cheaper for them to build more parking then it would be for them to provide effective connecting bus service. In some cases other transit providers do not provide good connections to rail – a good example is the DASH Studio City route not going to North Hollywood or Universal City Station even though it goes fairly close by both of them.
Re Metro stations are too large – I like the size of the Metro stations. Not sure we should hold up New York’s cramped, pre-ADA stations an example of a station design worthy of emulation.
“A Park-and-Ride that fills up at 7 AM is effectively one that doesn’t exist over much of the time period when it’s supposedly needed.” – It’s actually worse. Park & Rides that fill up by 7 (or 8) remove trips during the less congested part of the early commute only to run out of space at the peak. Drivers unable to find a parking spot are left to join the peak congestion. You can see this effect on Sound Transit’s 550 Express Bus at the Mercer Island and South Bellevue Park & Rides. Service levels into Seattle peak at 5 minute headways for the “get a parking spot peak” from about 7am-8:15. After that time, both Park & Rides are generally full but morning traffic keeps building until 9am or even later.
The description of “park and ride” here seems completely different to the way it exists and is used here in Britain. Over here, “park and ride” is usually used in relatively small towns, and refers to car parks at the edge of town linked to the centre by frequent bus services. It’s usually the most practical choice because the town centres have effectively medieval street layouts where everybody who has to work/shop in the town driving their car in would lead to horrific congestion, while the town centre’s shopping/working catchment areas in the more rural surrounds are so low-density that no effective public transport service would be economically justifiable. See for example the two places I’ve lived that have done this, Oxford and Shrewsbury.
I agree with Jarrett’s take on this, but I also agree with his comment that the provision of car parking near outer suburban commuter rail stations is an exception. In many of these cases population numbers and densities are both too low to sustain any sort of feeder bus service.
While some questions should be asked about how and why development was allowed to spread so far past the walk-up catchment in many of these areas in the first place, in most cases this is now a fait accompli. There is another problem with discouraging parking at stations in these outer areas; since most employment is also dispersed across these regions and often comes with free parking, stations without parking would be placed at a competitive disadvantage.
The issue is deciding the stations for which parking is appropriate and those where it isn’t. I guess a starting point is that stations which are destinations in their own right are less likely to be suitable, along with origin stations which have good feeder bus services.
This is a very relevant topic in Melbourne which is embarking on a series of redeveloped suburban rail stations to achieve grade separation on busy roads. The current approach is that any existing commuter parking is required to be replaced as part of the redevelopment. This is effectively precluding high value land being put to better use. Often a low grade car park is being replaced with an expensive new landscaped bitumen car park for free use by commuters when it would be better put towards mixed use or residential developments. Great opportunities for integrating new density and activity in transit rich locations seem to be being deferred indefinitely, or missed due to the “silo” nature of the the key project participants and a resistance to change by local government representing their communities.
Transperth introduced paid parking at all stations mid-way through last year, and now bays are available throughout the day, as well as increases in cycling to train stations: https://au.news.yahoo.com/thewest/a/25824472/paid-parking-eases-station-jams/
Interesting discussion regarding about a Park-and-Rides inability to address peak hour traffic. I would think there is still a value to having them located in outlying areas if one figures in the value of property in the downtown area but it would still be a subsidy favoring one mode of travel. I doubt, however, that was the motivating factor in the development of the Park-and-Rides in the Vancouver, WA area.
This article is right about the need for parking to be priced, but the negative externalities of the parking were not mentioned.
Park-and-ride induces sprawl and traffic when, really, the fundamental purpose of dedicated-guideway transportation systems is, and should be, to centralize the development pattern and to reverse sprawl.
The sprawl argument is one I’ve heard often. While there may be truth in it, it’s also worth considering the folks in your outer suburban areas, using large park-rides with 5 minutes express service to the core, that are brought into the community of transit advocates.
Great article
One minor correction – the photo is the park and ride at Albany, not Constellation. But yes it is on Auckland’s Northern Busway.