One justification for municipal minimum parking requirements is the danger of "spillover parking": the fear that if Big Brother does not force businesses to build huge parking lots, that business's customers will "spill over" into neighboring businesses or residential neighborhoods, thus reducing the parking available to the latter group. For example, if Wal-Mart doesn't build a thousand parking spaces, maybe Wal-Mart's customers will park at Mom'n'Pop Groceries down the street, thus reducing the parking available to Mom'n'Pop customers.
I recently noticed one flaw in this argument: spillover parking may be as likely to create positive externalities [that is, benefits to persons not party to a transaction] as negative externalities. For example, yesterday I needed to go to both the library and a bank. The library is a five minute walk from the bank, yet both have their own parking spaces. As a result, I had to drive to the library, then drive separately to the bank, thus creating an extra car trip and creating a tiny bit of pollution and congestion on the street between the two buildings.
By contrast, if I had just parked at the library and then walked to and from the bank, I would have been committing the sin of spillover parking: parking at building A while doing errands at building B. Yet I would have actually created positive externalities: by walking from the library to the bank, I would have reduced the amount of vehicle trips in the neighborhood, thus making life slightly easier for other drivers, and making the neighborhood air a tiny bit cleaner. On the other hand, I would have created negative externalities only if my car had been the last car to park at the library or bank, thus inconveniencing customers of one of those businesses.
Conversely, the parking requirements designed to prevent spillover parking have negative externalities as well as positive ones. Such regulations prevent the inconveniences that may result from spillover parking- but they also subsidize driving by forcing every business to build parking lots, thus creating an artificial glut of parking, thus driving the market price of parking down to zero, thus creating additional driving, thus creating additional pollution and congestion. And because these regulations encourage businesses to set their buildings back behind a wall of parking, they create an urban atmosphere that is not tremendously welcoming for pedestrians. Where shops are far from the street, pedestrians have longer commutes to their destinations, and those commutes are more dangerous as well because they must dodge vehicles on the way to shops and apartments.
In sum, the externalities caused by spillover parking may be as likely to be helpful as harmful- while the regulations designed to prevent such parking actually create harmful externalities.