Slashdot Mirror

Elon Musk is a smart guy, but he's pricing charging-equipped parking spaces below market equilibrium and then wonders why people sit in them all day.

All he has to do is install sensors and price each charging-equipped parking space a variable rate that maintains a roughly 15% vacancy at all times, like what San Francisco does. This encourages turnover and serves the maximum possible number of people.

It's like the tunnel he wants to build to bypass Los Angeles traffic. What will he do when that new tunnel gets congested? Is he going to block commercial drivers from using it as well, or is he going to build another tunnel below that one?

--Not Elon "It's Tunnels All the Way Down!" Musk

You have very different goals between private corporations and governments.

On this point we agree. For example, a politician thinks the optimal amount of parking is the amount that ensures parking is abundant even when the price is zero. The politician ignores the cost to the business owner and ignores the lost tax revenue from the unproductive land set aside for parking, and creates an unfunded mandate that the business owner must provide this parking at their own expense. By requiring overbuilt parking lots, the politician has thus added an unnecessary law to the books (read on to learn why it is necessary) and must also make up the tax revenue shortfall some other way.

On the other hand, a business owner thinks the optimal amount of parking is when the amortized cost of one additional space equals the revenue it would bring. If the parking lot is unpriced, sometimes it will get completely full and this will turn some customers away, but sometimes that's cheaper than building more parking.

A government run like a business would look something like SFPark where on-street parking is priced according to location, time of day, and day of the week to keep all the parking spaces about 85% full at all times, similar to the way an airline prices seats up and down according to demand. This completely and permanently eliminates the need to force business owners to provide parking, it clears room within city borders for more businesses to help share the tax burden, it provides parking revenue to the city that owns that land, it creates demand for parking garages and transit (which helps make transit less of a drain on the city budget), it keeps the flow of customers to the business relatively constant throughout the day and the week, and it prevents traffic congestion from people looking for parking and from double parked cars.

So you see, wonderful things happen when market forces are allowed to work in government.

At only 50 cents an hour on certain blocks at certain times, San Francisco has some of the cheapest pay parking in the nation.

it's very possible to build in a utility right of way, it just makes accessing the utility lines and pipes below the building impossible

That's not a zoning issue, that's a right-of-way issue. Anyway, why can't those utility lines and pipes be relocated at the developer's expense?

it's illegal because the dangers may not be completely obvious to the poor

So instead of solving the problem by requiring disclosure, you prefer to solve the problem by prohibiting the sale. Whenever a market failure arises, is eliminating the market always the best solution?

Got anything indicating that's happened in the last 45 years?

Yes, zoning laws are having a similar effect today.

someone who [wants off-street parking] will be more than happy that such a home was built for them.

So governments should force developers to build properties that have the widest appeal?

And those spaces can be in the driveway.

Where land is free and untaxed and the driveway is a dirt path, the driveway won't add anything to the cost of the house. But I don't think such a place exists.

street parking is woefully inadequate in a building with 450 apartments and only a dozen street parking spots in front of it

Only if those spots are priced below market equilibrium. If they are priced right at or slightly above equilibrium, then by definition there's no shortage of parking spaces and therefore no need to build hundreds of parking spaces for 450 apartments. Unless, of course, the developer thinks there's enough demand to justify the cost, then he or she will build parking without being forced to. (Trust me on this.)

Why have the car park itself at the mall at all? Why not have it head home so someone else can use it while you shop?

That's going to increase traffic congestion.

Anyway, we have already solved the problem of finding parking, and the technology is much simpler.

Then, 6-8 months later, the same people who said "there is no way I'm paying 10 quid for the privilege of driving in this congested city", begin to rationalize the cost, and start driving again.

So the obvious thing to do is raise the congestion charge.. Brilliant! Now rinse and repeat...

The concept works quite well for parking in San Francisco. There are no wild swings like the kind you describe because they don't raise or lower the price more often than once every 6 weeks, and then they don't raise it by more than 25 cents an hour or lower it by more than 50 cents an hour. Here is the pricing data data. Notice how the prices get more and more stable (the percentages next to the yellow circles) as time goes on.

One way to make the prices more stable more quickly would be to use what engineers call a PID loop.

Then, 6-8 months later, the same people who said "there is no way I'm paying 10 quid for the privilege of driving in this congested city", begin to rationalize the cost, and start driving again.

So the obvious thing to do is raise the congestion charge.. Brilliant! Now rinse and repeat...

The concept works quite well for parking in San Francisco. There are no wild swings like the kind you describe because they don't raise or lower the price more often than once every 6 weeks, and then they don't raise it by more than 25 cents an hour or lower it by more than 50 cents an hour. Here is the pricing data data. Notice how the prices get more and more stable (the percentages next to the yellow circles) as time goes on.

One way to make the prices more stable more quickly would be to use what engineers call a PID loop.

Better solution: Mandate that businesses (and by that, I mainly mean developers) provide enough parking for their projected number of customers and employees

Do you mean build so much parking that there's never a shortage when the price is zero? That's actually a very bad idea, because the economically optimal number of parking spaces is the number where MR=MC. This means wherever the cost of providing a parking space is not zero, the lost revenue from not providing it should also be nonzero. In other words, in an unpriced parking lot, it's financially optimal for it to get filled up completely at times. For a very similar reason, if you never miss a flight, you're spending too much time at the airport.

No, rather than micromanaging the number of parking spaces, it's better to decide what is the problem you're trying to solve, and give the businesses freedom to decide how to solve it. Is the problem that their parking lots are filling up completely? Then simply require that their parking lots never fill up completely.

It is either a gimmick, and won't catch on -- or it's going to require the entire infrastructure to be rebuilt around it, and won't catch on.

The way to go about doing this (if it requires an entire infrastructure) is to find one small piece to start with, that will be moderately profitable. That is what Nest did, instead of building an entire internet of things, they built a designer thermostat. Then they built a fire detector (which apparently doesn't always detect fires, but hey, modern technology). Their goal is to eventually build an entire IoT. My hope is they won't destroy the world with their insecure programming, but different people have different goals.

Same thing here. If you try to build an entire new infrastructure from scratch, you will fail. So start with smart parking meters.

While this is a great idea, in some cases it'll be a race to get an open spot, even worse than now.

They price the parking spaces according to demand in order to make at least one parking space available on every block. So there's no need to race to any spots.

Oh, and since the summary inexplicably didn't link it, SFpark is here.

In San Francisco, it's not in many locations (at least depending on the size of day), we have something called SFPark. http://sfpark.org/ Garages in New York and SF that I've visited also have similar policies and charge more for SUVs etc.

We have a similar system in San Francisco:
http://sfpark.org/how-it-works/