Urban Returns

The High Cost of Free Parking

I'm reading The Tome.  My favorite lines so far:

When planners set minimum parking requirements, they do not define demand and supply the way economists do.   For example, economists do not define the demand for food as the peak quantity of food consumed at free buffets where overweight diners eat until the last bite has zero utility.  Nor do economists, when asked for policy prescriptions, recommend that restaurants should be required to supply at least this quantity of free food no matter how much it costs.   Yet planners do define parking demand as the peak number of spaces occupied at sites with free parking, and cities do require developers to supply at least this number of parking spots, whatever the cost.

Cross-posted at Austin Contrarian.

Self-parody

Sometimes I think Randal O'Toole is putting us on.  He reacts to Frumination's calculation of the highway and parking capacity Manhattan would need to replace it's rush-hour subway capacity.  O'Toole's alternative?  Buses, of course:

Many 40-foot bus can carry 64 passengers (42 sitting, 22 standing, which means a higher proportion sitting than on a subway). Spaced five bus lengths apart, 11 buses per minute can cruise down a highway lane carrying more than 42,000 people per hour. That means fewer than 10 new lanes would be needed to carry the people now taking subways — and those 10 lanes would take up a lot less space than the 22 subway lines.

This can be further reduced by assuming that some people will time shift. During the 8 to 9 am hour, the average subway car carries 108 people, but in the afternoon peak hour the average subway car carries only 87 people. If some of the 108 people in that morning hour time shifted to the 7 or 9 am hour, then even fewer bus lanes would be needed to carry people into Manhattan.

Randal should have finished his calculations.  Forty-foot buses can cruise down a highway at 11 buses per minute only if they are averaging almost 30 mph.  But buses traveling into Manhattan could not approach that speed because once they got to Manhattan, they'd have to exit the freeway onto crowded local streets.  The ensuing bottleneck would back up bus traffic for miles.

Worse, Randal ignores the sheer amount of space his buses would occupy once they got to Manhattan.  It would take 6,250 buses to move 400,000 commuters.  At 40 feet per bus, 6,250 buses would take up 47 linear miles of street space -- if they were lined up nose-to-tail.  (94 miles if we allow one bus-length between buses.)  Manhattan is only 13.5 miles long.

And this ignores that commuters would need another bus network once they got to Manhattan, because express buses couldn't drop commuters at their final destination. 

This is absurd.  Randal must understand this, because he generously allows that buses might not be able to replace the entire subway capacity:

If there were no federal subsidies to New York City transit, it is possible that the subways would decline and not be entirely replaced by buses. If so, downtown Manhattan might lose some of its allure as a job center. Would that be so horrible? A lower-density Manhattan might have had less attraction as a terrorist target. It would save taxpayers money. And people would get to their work faster, as (thanks largely to its high proportion of transit commuters) New York has the longest average commuter times of any urban area in the nation.

It sounds to me like getting rid of the subway might benefit everyone except, of course, downtown property owners. And that is really what rail transit is all about: transferring wealth from millions of ordinary taxpayers to a few downtown property owners, rail contractors, and transit agency officials and employees.

Let's ignore that highways and bridges are also a subsidy to property owners.  (Is he factoring these capital and maintenance costs into his bus fares?)  Let's also ignore that all of the boroughs and suburbs served by the subway are much more valuable thanks to their accessibility to Manhattan. 

It's clear that Randal simply doesn't understand the concepts of increasing returns and agglomerations.  New York metropolitan residents are willing to endure longer commutes, pay higher commercial rents and housing costs, and fork over a good chunk of their income in taxes because they receive at least that much value from living in or close to the city.   New York City is a wealth-generating machine, and that is largely -- if not entirely -- due to a bunch of highly productive people crowding together.  Density begets productivity, and rising productivity is ultimately the source of rising wealth.   Turning Manhattan into Jacksonville or any other low-density city would destroy vast amounts of wealth.  Subway subsidies are trivial by comparison.  Most New Yorkers understand this, which is why you don't hear them argue for the elimination of the subway but instead leave that to an "economist" from Oregon.

(Cross-post at Austin Contrarian.)

Links

1.   Frumination works out the highway and parking capacity Manhattan would need to replace its morning rush hour subway capacity:

At best, it would take 167 inbound lanes, or 84 copies of the Queens Midtown Tunnel, to carry what the NYC Subway carries over 22 inbound tracks through 12 tunnels and 2 (partial) bridges. At worst, 200 new copies of 5th Avenue. Somewhere in the middle would be 67 West Side Highways or 76 Brooklyn Bridges. And this neglects the Long Island Railroad, Metro North, NJ Transit, and PATH systems entirely.

Of course, at 325 square feet per parking space, all these cars would need over 3.8 square miles of space to park, about 3 times the size of Central Park. At that point, who would want to go to Manhattan anyway?

See the neat map at the bottom of his entry showing how much land Manhattan would have to devote to parking.  Commuters of course would abandon Manhattan long before things reached this state.  But Manhattan would not be Manhattan in either case.  

Pointer from a reader.

(NB: Tyler Cowen thought this was interesting enough for a link . . . to Kottke, who merely quoted a snippet of Frumination's entry.  Jerk.  (Cowen, not Kottke.))

2.  A Detroit neighborhood reverts to prairie (images from DetroitYES):

(H/t the Urbanophile.)

3.  Professor Chris Nelson, Director of the Metropolitan Research Center at the University of Utah, projects  that the homeownership rate will fall to 63.5% by 2020, sparking intense demand for rban infill and redevelopment.   (Via Calculated Risk; no link to paper available.)

I think our homeownership rate is too high, a result of tax subsidies and land-use policies.  It ought to drop.  But it won't drop by several percent.   Richard Green, a professor at the Marshall School of Business at the University of Southern California (and no conservative), fisks Nelson's study. 

The new parking doctrine

For self-declared free-marketeers, we Americans have developed some deeply socialistic tastes.  We simply can't abide rationing some things by price.  Water.  Roads.  Parking.  Roads especially, but especially parking.

San Francisco's Livable City has put together a nice chart comparing standard American parking doctrine to a more market-oriented, or at least economically rational, doctrine:

Here's a fun exercise.  Go through the chart and replace "parking" with "roads."  The chart remains just as valid.   

Superficially, there is one significant difference between roads and parking:  government is the exclusive supplier of roads (except for subdivision streets); private developers supply most parking.  The market must be "choosing" the efficient amount of parking.  But this difference is an illusion, as everyone understands by now.  The government uses its own money to build too many roads or, more commonly, roads with too much capacity.  It uses minimum parking requirements to force private developers to build too much parking.  In fact, since it costs the government much more to build excess road capacity than to mandate excess parking capacity,  we should expect a greater oversupply of parking than of road capacity.

(Cross-post and comments at Austin Contrarian.)

Overparked

There are many reasons "too much" parking is bad for a downtown:

1. Parking raises the cost of new development, which means less of it.  This may be no big deal  for a city with a built-out downtown, but it is a big deal for Austin, which devotes so many downtown blocks to surface parking or stand-alone garages.
2. Parking not only raises the cost of new development, but it limits their size and density.  An on-site garage can only be so big to be practical.  A developer who wants to provide enough on-site parking to cover peak demand must first figure out how much parking he can build; only then will he know how much he can build of whatever it is he wants to build. 
3. Parking garages and surface lots blight the streetscape, triggering a negative feedback loop:   the surface lots and garages make streets less attractive to pedestrians, which drives the pedestrians away, which reduces demand for pedesestrian-oriented retail, which makes the streetscape even less attractive for pedestrians, etc. 
4. Subsidized parking -- i.e., parking provided below cost  -- distorts the market, encouraging an inefficient mix of driving and transit use.
5. Plopping ever more parking dowtown increases congestion.  The amount of land devoted to streets is fixed.  The amount of parking is not.  Increasing the number of parking spots but not the amount of street space means more cars per square meter of street, which in turns means more  congestion.   (This very interesting paper (pdf) by Michael Manville and Donald Shoup explores this argument in depth.)
6. Parking garages and surface lots are butt-ugly. 

There is one and only one counter-argument, but it should be taken seriously:   because we are so car-oriented, having too little parking will simply deflect businesses and shops away from downtown.  No matter how nice the sidewalks or inviting the buildings, a street is not pedestrian-friendly if it has no pedestrians.

I'm always willing to consider more evidence, but I don't think this argument holds for Austin.  There are good reasons to believe Austin's downtown parking market suffers from market failures.

But an even simpler analysis is to compare Austin's dowtown parking supply to other cities'.  If other cities get by with less, perhaps Austin could, too. 

Peer pressure

Last week's Newsweek featured yet another high-speed rail proposal, this one by America 2050:

I've discussed high-speed rail before.   Whatever the average Texan thinks about its feasibility in Texas, this map is likely to provoke a uniform reaction:   Why the hell should the Midwest get high-speed rail get before the Texas Triangle?   America 2050 proposes linking Chicago, Detroit, Minneapolis and St. Louis ten years before Houston and Dallas, and twenty years before San Antonio and Austin.  This, even though the Texas Triangle is more compact than the Midwest and growing at a much faster rate.

Although high-speed rail is taking a long time to get off the ground (so to speak), once a region  starts actual construction a lot of the doubts elsewhere will vanish under the pressure to remain competitive.  Even if most Texas are skeptical that HSR will produce much economic benefit, they will not want to gamble on being wrong, particularly since it would take 10-20 years to catch up.  The construction of a real HSR line in California or in the Midwest will likely spark a nationwide race for rail. 

On a related note, Tyler Cowen discussed Texas HSR a few weeks ago.  (He's mostly skeptical.)

H/t The Urbanophile.

(Cross-post and comments at Austin Contrarian.)