I didn't see Tyler favoring forcing prices above marginal cost, just opposing laws requiring excess supply.
So, there are two issues.
a. How much land should be devoted to parking spaces?
b. Given the answer to (a), what should be the price for parking?
I argue that for (b) the answer is often zero. A higher price would simply result in unused parking places, which does not increase welfare. Robin is falling back on issue (a), and here the thinking is that the state provides, either directly or through regulation, more parking spaces than are optimal.
Suppose there were no state provision of parking places. What would the equilibrium look like? Some possibilities:
1. You get Berlin, where the public transit is highly efficient and lots of people ride bicycles, even in the rain.
2. Individual housing developments and businesses undersupply parking. The thinking is that if parking runs out in front of your business, your customers will use the parking spaces in front of the business next door. This leads to stores putting up warning signs that say, "unless you patronize my store, your car will be towed." Neighborhoods put up signs that say, "unless you have a residential permit, your car will be towed." This imposes all sorts of enforcement costs as well as inefficient use of space. The warning signs often deter people from parking in places where they impose no cost at that particular time.
3. Land use responds, but not toward the Berlin scenario. On the contrary, businesses relocate farther away from cities, to locations where parking is cheap to supply and you don't get into fights with other businesses about towing rules. Housing developments are built without street parking but instead with large driveways--in effect, each household requires its own oversized parking lot to accomodate its peak demand . As a result of these sorts of adaptations, it takes more parking places to accommodate the same number of cars.
4. After a lot of Coasian bargaining, businesses agree to each provide a minimum number of parking places and housing developers agree to provide streets wide enough to allow parking.
The point is, you don't necessarily get (1). And you might get (4).
[UPDATE: Tyler's response to my earlier post. I still don't think he does enough to distinguish marginal cost from fixed cost. He also points to Randal O'toole, who questions whether minimum-parking ordinances are such a big factor.
Much to planners' annoyance, many developers elsewhere routinely provide more parking than zoning codes demand.
Yes, I am inclined to doubt that the amount of parking spaces we observe reflects a huge government-caused distortion. I worry that Shoup's goal is a huge government-caused distortion to limit parking availability.]