The costs of free parking

In the NYT, Tyler Cowen argues the case against free parking. It is a simple and compelling point. The way it is put in the NYT piece has created confusion (for example, see Arnold Kling).

If developers were allowed to face directly the high land costs of providing so much parking, the number of spaces would be a result of a careful economic calculation rather than a matter of satisfying a legal requirement. Parking would be scarcer, and more likely to have a price — or a higher one than it does now — and people would be more careful about when and where they drove.

He couches it in terms of all of the mandated parking and subsidised parking that is providing freely alongside roads resulting in over-supply but actually, it is just a consequence of congestion and environmental concerns. Put simply, every time someone provides and prices a parking space, they are failing to take into account the externalities this causes. That is, parking is a complementary service to driving and so is underpriced. Each additional spot causes more driving and more congestion — neither cost is internalised by the parking provider. So this isn’t a regulatory thing as much as an overall environmental issue.

But is it as easy as putting a price on parking as Cowen advocates? I’m not sure. While San Francisco is moving in a good direction in terms of congestion pricing on meters, it strikes me that usage based pricing is a better way to go — that is, taxing petrol, road use or emissions. But I do wonder how the free parking issue interacts with such policies. Sorting that out will require a model; something I don’t have time to undertake in this post.

5 thoughts on “The costs of free parking”

  1. It might just change the mix of stores and increase the number of freeloading stores that benefit from another store’s parking spaces. For example, a card shop (birthdays, anniversaries, etc.) might open up next to or very near a bakery known for its cakes. The first store to open would pay for the parking and the second store would benefit without additional parking spaces.

    Consumers purchase many items as a bundle at separate stores and the stores might open near each other. Stores selling items that are more expensive might open to minimize parking costs per purchase. Stores might also congregate to minimize search costs, such as several different women shoes or clothing stores near each other.

    Increasing the number of stores shopped per parking space would lead to increased time per space per vehicle. It would lead to more traffic congestion and not less, as the search time for a space increases.

    Parking cost, including search time for a space, is a transaction cost. It is irrelevant whether the storeowner or the shopper pays, as both will bear part of the final cost. When a transaction cost increases, suppliers (sellers) and buyers will look to minimize costs per shopping outing. The end result will depend on land use zoning, consumer preferences, sellers’ innovations and process improvements, etc. Once set in motion, the only sure is that the outcome will include many unexpected and unintended results.

    I seriously doubt the benefit, if any, will be as great as expected for congestion and environmental concerns. It probably will just change types of purchases and consumers.


  2. There are several distinct externalities to adding vehicles to the roads that need to be considered:

    congestion costs
    accident costs
    urban amenity costs
    “local” pollution (particulates, ozone, noise etc).  This is actually a huge cost, but gets ignored most of the time.
    greenhouse emissions.

    These have very different effects depending on location, and only a fairly complicated road use pricing model could properly account for them all.


  3. As Cohen notes their is a wastage from the wasted time and additional pollution from drivers cruising around seeking out the elusive (available) free park.  I wonder what the minimum number of such parks need to exist in a given neighbourhood for such behaviour to occur.  Or, put differently, if the number of such parks are reduced, what is the number below which most/all drivers stop searching and head for paid spots?
    It’s not particularly clear from the San Fran example how pricing is signalled to prospective parkers.  Surely if a driver must park before knowing the cost this raises the inefficiency issues (if drivers then chose to head elsewhere).  Having said that, my personal experience is that such unknown price situations are commonplace for anyone in a unfamiliar area, and most drivers would then rather pay the ‘going rate’ rather than ‘shop around’.


  4. It depends on how often a car park is used.
    In Melbourne, land averages around $1000 per square metre. An off street car park requires about 30 square metres. At 7%, this gives a per car park cost of $2100 per year or less than six dollars a day. If a supermarket averages six uses a day per car park, it obviously makes sense to provide those car parks.
    On the other hand, the Victorian government is busy building extra car parks for train stations. The average use per day would be about 0.8 (empty on weekends). This gives a per-commuter price of $7, which is more than the average ticket price. It gets worse. It could instead build (car park free) apartments on the land. This would save it the $30,000 and still give it at least one commuter, more for multi-storey construction. So the money spent on parking actually reduces the number of train commuters.. An example of (unfortunately popular) economic lunacy by government.
    The good news about the high price of private transport is that if we ever have to go without cars (peak oil, etc), we’ll have plenty of money freed up to build public transport.


  5. Professor Gans, I don’t understand why you say that “actually it is just a consequence of congestion and environmental concerns” and “this isn’t a regulatory thing as much as an overall environmental issue”. It is, of course, both.
    Really, there are three separate issues when it comes to parking. First, there’s the efficient rationing of a given stock of car parks, and the underpricing of parking by councils (and private operators—Westfield Bondi Junction I’m looking at you!), leading to queuing (e.g. cruising for parks). Second, there are minimum parking requirements for shops, offices and homes which force developers to provide more spaces than they would like, for which there is no market failure justification. Third, there’s the fact that parking is obviously associated with driving, and thus regulating parking can be a useful second-best approach to tackling the congestion and environmental costs of driving, given that more direct approaches seem to be politically unpalatable.
    Even if the third issue didn’t exist or was tackled directly, the other two would still remain. I don’t understand what you mean when you write, with regard to San Francisco’s smart parking meters, that “usage based pricing is a better way to go: that is, taxing petrol, road use or emissions”. The smart parking meters are designed to tackle the first issue, not the third. Of course, these alternatives are better means of addressing the third issue. But they are not very effective substitutes for addressing the first issue.  
    By the way, am I the only one surprised to see Tyler Cowen writing this stuff? Is it some kind of penance?


Comments are closed.

%d bloggers like this: