Sunday, May 31, 2009

Performance parking

Matthew Yglesias here argues for pricing public parking to match demand. There is certainly something to be said for this but Yglesias weakens his case with this:

... When you price street parking properly—which is to say a price that’s high enough so that there’s almost always a space or two free on every block, but low enough so that there’s not more than a space or two free on any given block ...

This sounds good but is just not practical, there is no way the pricing would be that attuned to demand.


  1. Yes way. Use cellular technology and electronic parking meters. This same technology, along with GPS has been incorporated into some advanced bus routes. In both cases, the inventory transportation needs, in time and location, is available to the user.

  2. Dynamic pricing schemes mean all the parked cars are not paying the same price which means the spaces are not being allocated to the highest bidders (which is what gives efficiency gains).

    For example you could price all the spots on a block at $0/hour until there was one left at which point the price for that remaining spot would jump to $1000000/hour. This would lead to there usually being one open parking space on every block but would obviously be inferior to a scheme in which every spot is free.

    If you never sell the last parking spot on a block you have effectively reduced the capacity by one space per block. But whenever you do sell the last space further arrivals will be locked out.

  3. "Perfect" pricing may never be possible, but surely many cities can do better than they do now. In many places, you can circle for 20 minutes without ever finding an open parking space, a strong indication that meter prices are much too low.