According to the lovely Dr. Elizabeth Taylor over at RMIT, parking in Melbourne’s city centre now occupies the equivalent of 225 MCGs.

The main reason for this bounty: minimum car parking requirements in the planning scheme. These requirements compel developers to provide a ratio of private car parks in all new developments.

Cost for land and construction ranges from $50,000 to $80,000 per 21m2 parking bay. To put that in perspective, the average one bedroom apartment is 45m2.

Surveys in Southbank and West Melbourne have revealed that between 26 and 41 per cent of private parking spaces are empty.

City of Melbourne planners are rightly calling for regulatory reform. The car park glut represents a misallocation of resources.

We’d be better off using that land for housing, or green space or something else currently undersupplied in the Melbourne CBD. Furthermore, autonomous vehicles and other disruptive technologies are predicted to reduce demand for car parking.

What about the cost of supplying all that extraneous land for car parks? If we reduce the cost of providing car parking, we’ll reduce the cost of apartments. Right?

Well, not exactly.

Mandatory car parking probably suppresses the value of land. This is because car parking is a relatively low value use (as indicated by widespread under utilisation).

Less land for apartments = less income = lower land prices.

It is unlikely that the size of inner city apartments will automagically grow when we stop squandering land on car parking. Rational developers will simply add more 45m2 apartments to increase their project’s overall yield.

Remove the minimum requirements and car parking must compete with commercial and residential land-uses.

More land for apartments = more income = higher land prices.

There is at least four factors at play: a regulated minimum requirement; an underdeveloped market for privately owned car parking; the ‘bundling’ of car parks into apartment titles; and, a highly competitive land market.

In Montreal, the city has removed planning restrictions to allow peer-to-peer rental of parking spaces. They are also allowing residents to rent their verge to a neighbour for community-based agriculture.

Melbourne can do like Montreal and enable a peer-to-peer market: an “AirBnB” for car parks. Were it to establish, this kind of market would eventually reflect the true cost of providing private CBD. Planning, body corporates and building security all make this tricky but not insurmountable.

Another suggestion is to reduce the public land subsidy to car parking by improving on-street pricing regimes. In San Francisco they use dynamic pricing (sort of like Uber’s surge pricing, but for car parks). In Tokyo your car must have a registered car park, and any “homeless” cars are towed off the street overnight. These regimes are designed to reflect the spatial cost of cars.

The car park glut exemplifies the way in which planning regulation can act like taxation on land. Mandatory inclusions are factored into the price developers are willing to pay for land. If they have to provide a car park, they’ll pay less for the land.

Now imagine that instead of requiring car parks, the planning scheme required social housing units. This is called Inclusionary Zoning 

Developers argue that the market cannot bear this, but in fact the cost capitalises into a lower land price. The low value use acts like taxation on the land.

If the planning scheme can produce a glut of homes for cars, I wager it can produce a glut of homes for the poor.

The only question is how to make it fair for developers who have sunk costs under the old rules.