Sunday, July 19, 2015

The perverse effects of on-street parking

I've been silent for quite a while, a result of my moving to a condo in downtown Montréal, but now it's finally (mostly) over, so back to our scheduled programming...

So let's get back to one of my biggest heresies with regards to conventional urbanism: namely my opposition of on-street parking.

Most North American and European urbanists tend to favor on-street parking. The reasoning is that cars parked on the street narrow the roadway and protects pedestrians from cars in movement. It also protects the pedestrian corridor from intrusion by cars. My counter-argument is that on-street parking claims the street for cars exclusively and pushes pedestrians on narrow sidewalks, squeezing them between building walls and steel walls. As streets are the most dominant public place in cities, parked cars result in them being taken away wholly for vehicles, in movement or parked.

I also believe that one of the first rules of good urban design is that we must treat cars as a shameful disease to be hidden from the public view so that urban design can be centered on human beings. The presence of too many cars is deleterious to any urban environment, so having cars right in your face, parked on the curb, is the exact opposite of what we should aim for. Hide the cars away so that you can focus the environment on people, not vehicles.

Compare and contrast:
Lyon in France, remember that this isn't the point of view of pedestrians, but of car drivers, pedestrians are pushed to the side
Sendai, in this case, pedestrians really have this view

Another place in Lyon, look at the lack of space between cars and building walls, is this some place to hang around in?

One residential street in Sendai
 Now look at these major arterial streets, again in Lyon and Sendai:
Arterial in Lyon

Arterial in Sendai
But again, this is a view from the street, not from the sidewalk, where people will actually walk, what do the sidewalks in each example look like?
Lyon, narrow sidewalks, where only two people can walk side by side, stuck between cars and walls

Sendai, sidewalks seem 2 to 3 times as large, providing a comfortable walking area no matter how many pedestrians there are, and even a bike path
Anyway, taste may be subjective, and while I much prefer the Japanese examples, some may favor the French ones. But that is not the point of this post. I'm instead going to think about the practical ramifications of on-street parking

The government as a parking provider

One of the most evident consequences of deciding to satisfy parking needs with on-street parking is making the government a main parking provider, if not the most important parking provider. Streets are almost the very definition of a public good, they can't readily be privatized, they must be planned and publicly provided. As a result, that means that the parking spots that can be found on the street are also publicly-owned. 

So when governments commit to build on-street parking everywhere, how much parking are we talking about here?

Well, if we take one average city block that is 300 meters or 1 000 feet long and 60 meters or 200 feet deep, and we have a public right-of-way that has 2 3-meter (10 ft) lanes, 2 2,5-meter (8 feet) shoulders for parking and 2 1,5-meter wide sidewalks (5 ft)...
Basic ROW shape
...that is a total of about 14 meters or 46 ft of ROW.

So the typical block would look a bit like this:
Typical block, in red, the parking places
We then have a 74 m x 314 m block size, including public ROW, for a total of 23 236 square meters, around 5,8 acres.

We have 720 meters of parking space on the street (2 x 60 m + 2 x 300 m). Supposing an average of 6 meters per car, that is 120 parking spots, which occupy about 8% of the land area.

So by committing to build space for parking on the street, the government commits to building 51 parking spots per hectare or around 21 per acre, or for a bigger scale, around 5 000 parking spots per square kilometer (13 000 per square mile).

So if you have a city with a  built area of 200 square kilometers (80 square miles), that city will have built around 1 million on-street parking spots. That is undeniably a massive amount of parking, all provided by the city government.

Ramifications of the government as a parking provider

The fact that the government is a parking provider has a lot of consequences. Since the government is providing so much parking, parking will be seen as a public service, because, well, it is a service that is provided by public authorities. People who depend on that service will therefore hold the government accountable for the quantity and quality of parking. In other words, more than just a parking provider, the government will be forced to take on the role of managing the parking supply in order to satisfy its voters.

This role of supply manager is actually what leads governments to adopt minimum parking requirements and limits on density (most of the objections to new developments in already dense areas are rooted in fear of overwhelming existing on-street parking, making it harder for residents to find parking there). Off-street parking requirements were conceived initially as a way to prevent parking overflow on city streets, in order to keep parking spaces free on the street, it was required of developers to build off-street parking so that the people who live/work/shop in these developments wouldn't overwhelm the public provision of on-street parking spaces.

Without on-street parking, the government could have left developers to deal with their parking problems by themselves, or let the market provide only as much parking as drivers were willing to pay for. With on-street parking, city governments were forced to intervene, as a major parking provider itself, it couldn't turn a blind eye on issues of parking supply and demand. The government can't say that parking issues aren't its business when it is the main parking provider in the city.

Of course, in a situation where on-street parking is significantly under-priced or even free, any off-street parking operator choosing to put a price on its parking lot can expect people to just park on the street if there is any place available there, which can result in parking overflow on streets even as the parking lot stays empty. This can lead to the city government coming down on that parking owner, because this is exactly the situation that the city government sought to avoid by having minimum parking requirements.

Economic ramifications

In case I've not made it clear enough, my position on parking is the following: there should only be as much parking as drivers are willing to pay for, parking supply should neither be inflated by government (minimum requirements) nor limited by government (maximum parking rules).

In general, letting developers decide how much parking to build should result in this "user payer" situation, as they will make economic calculations and build only as much parking as they can justify economically. But what happens when they have to compete with on-street parking, often provided for free by the government? It completely distorts the market, private off-street parking options will generally not even exist, because who would park there for a fee when they could park for free on the street? At least, not unless on-street parking is completely overwhelmed.

As a result, on-street parking completely destroys the pricing mechanism for parking. Donald Shoup favored a pricing ideal for on-street parking in that the price of parking should be such that 80% of parking spots are occupied, if it's more, rise the price, if it's less, lower it. That is one approach... but implementing it still means that much (if not most) of the time in most locations, the proper price to reach that goal is... 0$. So you still have people being provided a parking spot for free in many places, especially in low-density areas. Yet on-street parking still has a cost, it's not free to provide.

The reason why the market can provide a good approximation of "user-payer" in many cases is that supply is free to increase or decrease with price. If the price of a good increases, then developers will provide more of it to profit from it. On the other hand, if the price of a good decreases, or if more profitable uses of resources come to be, supply can decrease as previously profitable parking spaces can become less profitable than alternatives, or even outright unprofitable.
The owner of a private parking lot has the option of continuing operating his lot or selling for development, in the case presented here, over a 30-year analysis, it's better to keep operating the lot...

But if the value of land increases, it may become more profitable to just sell the lot for development, reducing the supply of parking but increasing the supply of land to develop

But with on-street parking, that's not the case at all. We are dealing with a fixed supply set up by the government. You can vary the price, but you cannot decrease the supply, at least, not in any useful way. One of the big problems is the geographic form of on-street parking. These narrow bands of asphalt between the travel lanes and sidewalk are utterly useless for developers. The best you can do is rearrange them to be wider sidewalks, bike paths or a green buffer between street and sidewalk, but nothing else. There is no versatility here, because the land cannot be recuperated for development, it doesn't have to compete with alternative, productive uses for it.

So if we agree that a proper parking policy should result in parking users paying for the parking that they use so that their transport choice isn't subsidized, then on-street parking is the greatest thorn in our side. It is hard to price, and the supply is fixed and cannot vary depending on demand, and as a result, in lower density areas, it will push out private for-fee parking and result in on-street parking and mandated off-street parking being free.


So there was my take on the perverse effects of on-street parking, how it makes providing parking and managing parking supply a government responsibility while disrupting the pricing mechanism for parking spots, largely eliminating for-profit parking businesses except in a few exceptional places. I accept that, unfortunately, most North American city streets are already built much too wide to accommodate parking, but I think that on-street parking is one of the greatest obstacles to a sane parking policy that would have drivers assume the full cost of the parking they use.

Parking shouldn't be a government's responsibility, yet that is ultimately the result of allowing on-street parking to be massively used. This leads to the absurd hypocrisy that city governments will take more seriously their "responsibility" to provide sufficient parking to all drivers, but not their responsibility to make sure all their residents can find proper housing. As if an human being not being able to afford a place to live in is sad, yet tolerable, but a car without a parking space is unacceptable.


  1. I agree with you in general but because we're stuck with all these wide streets, there's not much we can do. It always bugs me when new streets are built with on-street parking bays included. Wasted opportunity.

    There's another obstacle to reclaiming the parking spaces as wider sidewalks: business owners generally fight to the death for 'their' parking spaces. You could propose to replace all the parking spaces by creating an off-street garage, but that's expensive, and made even more difficult by having wide streets in the first place. The on-street parking that was created by cars squatting on existing streetscape is cheap, already there, and hard to push aside for political reasons.

    By the way, places like Brookline, MA ban overnight on-street parking (effectively), so presumably you have to find yourself an off-street parking space if you own a car there. Unfortunately, the Brookline zoning code still has onerous off-street parking minimums, even though there's no 'spillover effect'. The old-timers are quite attached to the notion, presumably for class or status-seeking reasons.

    And there is actually spillover, but it happens across the border in Boston, where decades of incoherent parking policy has resulted in many blocks having unregulated parking, other than street-cleaning. The result: many Brookliners (and other suburbanites) dump their cars on Boston streets overnight or during the day. Then we get the twisted outcome of Boston residents demanding that off-street parking be supplied in new Boston developments "because all of the on-street parking is occupied" (by outsiders!). New Boston developments effectively subsidize the free parking of suburbanites. It's quite amazing...

  2. I'd like to zoom in on one comment: "most of the objections to new developments in already dense areas are rooted in fear of overwhelming existing on-street parking, making it harder for residents to find parking there."

    It's true that most middle- and upper-middle-class NIMBYs in central cities frame their objections in terms of parking, but the universality of NIMBYism along all classes and forms of development suggests that this is an excuse. In some cases, the NIMBYs are open about what they want: see, for example, the "apartment dwellers and renters are yucky" comments made in richer neighborhoods, e.g. the examples recently quoted by Daniel Kay Hertz. In other cases, the neighborhood doesn't even have high car ownership, but the NIMBYs either oppose any road diet that they perceive as coming from outsiders (Washington Heights) or explicitly say they don't want new people who they don't identify with (the Mission).

    But for a moment, let's suppose that the NIMBYism really is about parking. Why would off-street parking change things? More development in the neighborhood would raise the price of parking. There would be more demand; very roughly, the price of parking would go up proportionally to land value assuming free development, as lot owners converted parking to buildings. Development restrictions are a nifty way of disconnecting land values from parking rates, and are especially useful since urban drivers usually own apartments but usually rent parking space, possibly from their own building manager.

    1. Not all NIMBYism of course is about parking, but that comes out often when people already use street parking a lot. It may be just a rationalization for some, but it is also a real fear for others. For example, one of the measures proposed that allowed the Roslyn-Ballston corridor development to beat opposition was making street parking around the area limited to residential permit holders, and giving those permits only to residents, reassuring them that only they and their visitors would be able to park in the street.

      Higher density development could indeed trigger NIMBY opposition for fear of rate increases. But I think rate increases are a lesser fear for people already paying for their parking than not finding any parking near their home, especially as people who own a parking spot on their property wouldn't be affected at all.

  3. Is $300,000 over 30 years really a winner over $150,000 up front? If you can get a 2.4% return on investment, then you'd rather take the $150,000 up front.

  4. One thing that's always bothered me about the support for on-street parking by urbanists is that it presumes the parked cars will always be there to provide that buffer from traffic. Reality isn't usually so simple. There's rush-hour parking restrictions, street cleaning times, no overnight parking, or in commercial areas a simple lack of need for much street parking at night. This is when these wide streets become drag strips, or during rush hours the worst kind of automobile sewers, but that's exactly when people on the sidewalks need the most protection.

    The visual and psychological impact is hard to deny too. I'd also make the comparison between a parking lot and a plaza. Who wants to look out onto a big parking lot? What about a big open plaza though? The only real difference is that one has parked cars in it and the other doesn't. Many plazas are also rather sparsely vegetated and monolithically paved, but the difference in having cars versus no cars is like night and day.

    1. Well, the pro-on-street-parking urbanists I know also oppose rush-hour parking restrictions...

      Bear in mind that my contact with urbanists is largely in very dense cities, like New York, or the densest parts of Boston, and there, 5,000 parking spots per km^2 can fill even at low car ownership rates. This means that most of the spots are taken, and even at less busy times, enough are taken that the parking lane is not a useful moving lane.

  5. The master stroke of the Japanese was to require every vehicle to have its own off-street parking space instead of requiring developments to provide spaces for which there may or may not be a car; this immediately creates a market for parking by establishing a demand, while leaving supply up to the market ensuring that only as much parking as needed is provided.

    This in combination with their people-focused street widths has made their cities some of the best in the world, despite the architecture being lackluster in many cases.

    1. Indeed, the "proof of parking" has been a genius move by putting the responsibility for finding parking on the prospective car owner rather than on government and planners. Japan still has some off-street minimum parking requirements in certain places, but the amount is really low, like one half or one quarter of that in North America.

  6. I agree with you it would be much better for our quality of life in the city if parking was not permitted on street.

    As we are all aware for the time being, this is only a pipe dream and the only way we can even dream of getting cars off the road is by having some major rethink on parking charges. In order to make developers build more off street parking, we would require major policy changes for new owners or occupiers to only allow them parking on street by having to pay at least a daily rate, but even better, would be an hourly charge.

    This would be seen by many citizens a major drawback to living in the city, but over a period of time as we see less and less cars on street, we would quickly see the benefits.

    The only way such a policy could have any chance of succeeding, would be by having national and local politicians fully behind it and in reality this would be an even bigger pipe dream.

  7. The 5000 spots/km2 is an interesting number; looking at urban population densities in the US, that's seemed about the plateau for cities (Boston, Chicago) that try to juggle transit (especially for going downtown) with widespread (and surface parked) car ownership, going as high as 7000 people/km2 for SF or Somerville. Vs. Brooklyn at 14,000.