I am disappointed that the distinguished George Mason University economist, Tyler Cowen, has fallen for the "high-cost-of-free-parking" arguments of UCLA urban planner Donald Shoup. Shoup is an excellent scholar, but like many scholars, he has the parochial view that the city that he lives in is a representative example of what is happening everywhere else.

Should free parking be a thing of the past?

Shoup's work is biased by his residency in Los Angeles, the nation's densest urban area. One way L.A. copes with that density is by requiring builders of offices, shopping malls, and multi-family residences to provide parking. Shoup assumes that every municipality in the country has such parking requirements, even though many do not, and that without such requirements there would be less free parking. This last assumption is extremely unlikely, as entrepreneurs everywhere know that (outside of New York City) 90 percent of all urban travel is by car, and businesses that don't offer parking are going to lose customers to ones that do.

Shoup portrays such free parking as a "subsidy" because not all people drive and so the ones who don't drive end up subsidizing the ones who do. But any business offers a variety of services to its customers and employees, and no one frets about subsidies just because they don't take advantage of every single service. How often do you actually swim in the swimming pools or work out in the exercise rooms of the hotels you stay at?

Shoup also supposes (and Cowen accepts) that universal parking fees would greatly reduce the amount of driving people do. "Minimum parking requirements act like a fertility drug for cars," Cowen quotes Shoup as saying. Metro, Portland's regional planning agency, submitted this question to its transportation model and concluded that requiring all offices, shopping malls, and multi-family residences to charge for parking would reduce driving by about 2 percent. The model showed that charging for parking has a greater effect on driving than spending billions on light rail, building scores of transit-oriented developments, or increasing the urban area's population density by 20 percent. But 2 percent still isn't going to do much to relieve congestion or solve any of the other problems Cowen associates with driving. Plus he never really explains why he thinks reducing mobility is a good idea in the first place.

Shoup claims that a single parking space costs, on average, 17 percent more than the cost of an average car, and as a result, the cost of parking greatly exceeds the value of all automobiles in the country. This is ridiculous. Most free parking is surface parking, which costs about $2,000 a space plus the cost of land. In areas that have not used urban-growth boundaries and similar tools to create artificial land shortages, vacant suburban land with urban services typically costs about $20,000 an acre. Since each acre can hold about 100 parking spaces, the total cost is about $2,200 per space. From the point of view of a business owner, this cost can be amortized over 30 years at 6 percent, for an annual cost of about $160. If that parking space is used by just two customers a day, the cost is about 22 cents per customer. That's pretty trivial, and the costs of collecting fees for such parking would probably be greater than the parking itself. Even structured parking typically costs only about $10,000 a space (or, using the above assumptions, $1 per customer), but structured parking is rarely provided for free.

Strangely, one of the examples Cowen uses in his article is Manhattan, where (he claims) "streets are full of cars cruising around, looking for cheaper on-street parking, rather than pulling into a lot." Give me a break! I defy Cowen to find any free parking anywhere in Manhattan, where ownership of a single parking space can cost more than a median home in other parts of the country.

Cowen's complaint about Manhattan is not about free parking but that the government is pricing on-street parking below the market. If that were the extent of Shoup's argument, I would have no problem, as I noted in my blog last week. But Shoup's goal isn't market pricing of public parking; it is to create artificial shortages of private parking. He doesn't want to simply eliminate the minimum-parking requirements that are found in many zoning codes; he wants to replace them with maximum-parking limits so that places like WalMart will not be allowed to provide their customers with as much parking as they like.

The empirical question is: do shopping malls, office parks, and companies like WalMart provide parking for their customers and employees because of zoning mandates, as Shoup claims? Or would they and do they provide parking just because it is good for their businesses? Texas counties are not allowed to zone, yet shopping centers and office parks in unincorporated Texas still provide plenty of parking. Much to planners' annoyance, many developers elsewhere routinely provide more parking than zoning codes demand. This suggests that free parking is a free-market choice, and Cowen, who generally supports free markets, should have no objection to it.