Urban Returns

"No road we built in Texas paid for itself."

2009-11-12T13:48:59-06:00

Over the past two days at the Congress for the New Urbanism Project for Transportation Reform conference, attendees have called for reform at local, regional, and national levels. In a panel debate about the future of transportation funding and the role of regional planning through MPOs, several speakers argued that the foundation of transportation and development funding had to be systematically overhauled.

Mike Krusee, [AC: former] chairman of the Texas House of Representatives Transportation Committee, said that financial problems were more significant than environmental, though they should be tied together in the same discussion.

"The reason there's not a new transportation bill is because there is no money. We've hit the wall of unsustainability on how we finance the transportation system," he said.

Krusee asserted it was urgent and necessary to understand the nature of this broken financial apparatus and to develop solutions to fix it. In Texas, he said that, on average, it cost the state 20-30 cents per person per mile to build and maintain a road to the suburbs, yet drivers only pay on average 2-3 cents per mile through the gas tax, vehicles fees, etc.

"What we found was that no road that we built in Texas paid for itself," said Krusee. "None."

The expense to build roads and utilities further and further from the urban cores not only drove costs to unsustainable levels, it created an imbalance in who paid for growth. Over the past 50 years, Krusee argued, the federal government used tax money that came by and large from cities to subsidize roads to areas without access otherwise.

"City dwellers have subsidized the land purchases and the development costs out in the suburbs," said Krusee. What's more, the gas tax, which city dwellers pay when driving on city roads, but which goes to freeways largely outside of urban cores, is "a huge transfer of wealth from the cities to the suburbs to build these rings."

Krusee said building the interstate system was initially a good thing, because if facilitated interstate commerce and increased the productivity of cities. Now however, because of congestion caused by ever longer commute patterns, system productivity is in peril. "What's happened is the federal government has basically reneged on the deal. By subsidizing highways out to the suburbs, it's no longer efficient for truck traffic, for goods and services and people to move between cities in the United States because those roads have been hijacked by all the commuters."

Krusee, by the way, represented suburban Williamson County.

Krusee's assessment matches TxDOT's own internal assessment. (This actually should be no surprise since Krusee's committee relied on TxDOT for data.) TxDOT, for example, concluded that the 15 miles of SH 99 from I-10 to US 290 will cost $1 billion to build and maintain over its lifetime, while only generating $162 million in gas taxes -- just 16% of the total cost.

Some of us get swept up in the rhetoric sometimes, but roads aren't unmitigated evils. Obviously, we need roads. Just as obviously, I think, we will continue to need new roads. But new roads should be built only where drivers are willing to pay for the new capacity. And the only way to gauge that demand is to price existing roads properly; the revenue they generate will tell us when it is time to add to add that capacity.

Cross-posted at Austin Contrarian. Comments there.

Aragon

2009-11-12T13:43:52-06:00

Mueller gets a lot of criticism -- see the grouchy comments to this entry -- much of it unfair. I like Mueller. I do agree, though, that it is not a model mixed-use, New Urbanist development. Among other things, there is too much segregation of single-family and multi-family/commercial.

So let me offer an example of a model New Urbanist development. Aragon is an infill development in Pensacola designed by Michelle MacNeil (who happens to be my cousin). It's several years old, but I got my first tour in August when we visited her during a trip to the Gulf Coast.

One of the most significant features of Aragon is its location. Many New Urbanist developments are suburban greenfield developments or massive redevelopments of abandoned industrial land or airports (Mueller and Stapleton, for example).

Aragon is a true infill development. It is just a few blocks east of downtown Pensacola and directly north of Pensacola's Seville neighborhood, which dates to the turn of the 19th century. One of the challenges, Michelle explained, was integrating the neighborhood with the Seville neighborhood, providing a smooth transition from the old to the new.

View Larger Map

Here is a shot with the god-awful convention center (aren't they all?) in the background.

Although every home in Aragon can be built-out as two-family, it is mostly single-family, which matches the density of the Seville neighborhood. But it does a better job of mixing different types of housing and other uses than Mueller, and much better than standard suburban tract developments.

Here are row houses lining one of the main streets:

These are true row houses, unlike the row houses at Mueller, which are actually four-plex condominiums. Each of these sits on a separate lot owned in fee simple. Lot owners build what they want -- subject to a detailed set of design criteria.

Here are the same row houses from the rear:

Allowing each home owner to choose his style gives the row a more varied, interesting appearance than Mueller's standardized four-plexes. (This photo was taken from the rear balcony of one of the most expensive houses in Aragon, by the way.)

Around the corner from the row houses are live-work units (sorry for the bad picture; the sun was shining right in the camera lens):

Top residential, bottom commercial/retail/office.

Parking is in the rear, connected by alleys:

The rear parking lot is . . . just a parking lot. Nothing special. But when parking is relegated to the rear, it doesn't have to be.

Speaking of mixed-use, all (or almost all -- I don't remember) of the homes can can be used as offices. This business sits on what is otherwise a purely residential street:

The architecture matches the Southern architecture of the old neighborhoods to the south:

This not tract housing, by the way. Lot owners can hire their own architects and builders but, again, must comply with detailed design criteria.

And the criteria are quite specific. These large homes lining a crescent park are required to be three-stories tall. (Note a couple of unbuilt lots.)

This green is lined with much less expensive homes.

These houses are actually quite small and close together. In order to minimize the sense of crowding, the architectural regulations require that eaves and balconies and porches line up precisely:

The porches, too:

The uniform lines trick the eye, preventing it from easily spotting where one house begins and the other ends. As a bonus, the "wall" of housing makes the green a cozier, more inviting space.

Another technique for creating the illusion of space is to make good use of the space between homes. Usually, this is where the air conditioner units go. But the homes along one street are separated from one another by side yards:

The neighborhood is a pleasant place to walk. It's got good sidewalks, nice green space, etc. But the sheer variety of housing, the fine mixture of styles and sizes, makes the place visually interesting. Almost every spot in the neighborhood provides such a vantage point; no block has the dull uniformity that's created by strictly segregated housing types. Unfortunately, that's hard to capture with a camera:

One thing good infill does is beget more good infill. This is a new multi-family development just down the street:

Here is the property to the immediate west:

This is one of Pensacola's oldest cemeteries. If you want a rough metric to separate true infill from faux infill, distance from an old cemetery is probably as good as any other.

Finally, this space at the southeast corner is reserved for a pure commercial and retail center:

(That's Pensacola Bay in the background.)

The space is still green, of course. Development has been stalled by the collapse of the credit markets. But attracting commercial development has always been the Achilles heel of New Urbanist developments. HEB and A&P and Walmart want their big asphalt parking lots. I think Aragon has a much better chance of seeing infill commercial development than most, though, because it's smack in the middle of the city.

Aragon has other fans. If you want to see more, here is a gallery of photos by one of them.

Comments closed. Cross-posted at Austin Contrarian.

Is it really necessary for the federal government to introduce new market distortions?

2009-09-05T15:37:25-05:00

For the reasons I just gave, I don't believe that our past subsidies to suburban development justify maintaining those investments. End them and let the chips fall where they may.

But even if reasonable people can disagree about that, I see no argument for creating new subsidies. But that is exactly what our federal government is about to do.

The FHA is a major player in the mortgage business. It insures mortgages for home buyers who otherwise would not be able to get credit or scrape together a down payment; this allows some to buy homes with as little as 3.5% down. Without the FHA guarantees, the mortgage market would be much smaller. A few years ago, its share of the market was 2.7%. This seemingly small share had an outsized effect by scooping up the biggest credit risks and reducing the risk for private mortgage insurers. In any event, its share of the market has ballooned recently: according to Friday's WSJ, its market share reached 23% in the second quarter of this year.

And now the FHA is about to adopt new rules explicitly encouraging home buyers to purchase single-family, detached housing in the suburbs rather than attached (condos) in urban cores:

Until now, almost any condo development could apply to FHA for “approved” status, therefore making FHA financing available in that development. In addition, in developments that were not approved, “spot approvals” were sometimes available for individual units. (The lender applied for an approval for the unit you wanted to buy, in spite of the development not being approved).
Following are the new guidelines: (This is not pretty, so prepare yourself)
1. There will be NO more spot approvals.
2. All development not considered primarily residential are out. For instance, a development with more than 25% of the total floor area dedicated to commercial business use is out.
3. Noise issues is a new concern, so any development within 1,000 feet of a highway, freeway, or heavily travelled road, 3,000 feet of a railroad, 1 mile of an airport, or 5 miles of a military airfield will become ineligible for approval.
4. If the property has an “unobstructed view , or is located within 2000 feet of any facility handling or storing explosive or fire prone materials, it is not insurable - we're not talking just fireworks factories here. A gas station 2 blocks away can disqualify this development.
5. Any property located within 3000 feet of a dump, landfill, or superfund site, is ineligible.
6. No more than 10% of the properties can be owned by a single investor, including builders or developers who are renting out or have not yet sold vacant units. For 2-3 unit developments, no one can own more than one unit.
7. No more than 15% of the homeowners can be more than 30 days late on their homeowner dues.
8. For new developments, at least 50% of the units must be sold prior to applying for FHA approval (valid presales include those with purchase agreement and lender validation of an approved loan in process)
9. A minimum of 50% of the units must be owner occupied or sold to owners who intend to occupy as their principal residence.
10. Projects in designated wetland and flood zones will not qualify.
11. All current condominium project approvals will be invalid (with the exception of projects approved on or after October 1, 2008) and projects must be re-approved under the new options available. Going forward, all projects will require recertification every two years.

Perhaps most insidiously, the FHA will no longer insure more than 30% of the units in a single condominium development.

I understand that the FHA has been taking large losses on its guarantees. I also understand that condominiums are riskier to insure than single-family homes. Some of these new regulations seem reasonable -- it's probably a bad idea to invest in a condominium when the condo association lacks the resources to pay for upkeep. Higher risks justify higher premiums, too.

But these new regulations seem purposely designed to push new homeowners out of dense, urban areas to the suburbs. They exclude many mixed-use developments (#2). In a central city, it is hard to find a condominium not within 1,000 feet of a highway, freeway, or heavily travelled road, 3,000 feet of a railroad, or one mile from an airport (#3). Allowing developers to tap into FHA guarantees for entire single-family subdivisions but only 30% of condominium units naturally will encourage developers to shift to single-family subdivisions. These new regulations are fundamentally anti-urban.

Even if it is somehow possible to defend our existing scheme of suburban subsidies, is it really possible to defend introducing new market distortions?

(Cross-posted at Austin Contrarian; comment there.)

Path dependence

2009-09-05T15:34:05-05:00

Ryan Avent has had some back and forth with Tyler Cowen and Will Wilkinson over roads and rail. Tyler and Will acknowledge that, yes, our reliance on roads rather than rail has been influenced by long-standing government subsidies. But by now we have sunk so much into roads, and have built up so much road infrastructure, that the marginal cost of adding to our road network is much lower than creating a viable rail network from scratch. In other words, our past investments have had a lock-in effect, and there's no wishing it away.

Ryan responds that Tyler and Will are committing the "sunk costs" fallacy, but I'm not sure that's right. Past investments matter a lot because they have generated network externalities. One million dollars spent extending an already extensive network is often worth more than one million sunk to build a network from scratch.

But Ryan is right that past investments in suburban infrastructure do not require us continue throwing up barriers to denser developments:

In practice, the US is far from done building. Tens of millions of new homes will be built in the coming decades. Hundreds of billions of dollars will be spent on transportation infrastructure. The current built environment has, as a result of decades of government policy, taken on a rather suburban, auto-centric tilt. So what? No one is suggesting that we tear down all of that and replace it with something entirely new. I, and others, are suggesting that making it easier (or, you know, legal) to build in a denser, more walkable fashion would be advantageous. Similarly, given the burden of maintaining such a large and costly road infrastructure, it might be wise to devote a larger share of dollars for new construction to substitute technologies.

And bear in mind that a technology's lock on the market is not always as secure as it seems. A small initial advantage allowed VHS to dominate the videotape market, all but squeezing out Betamax. But it did not save VHS from the DVD, even though a vast infrastructure had been created to support VHS: video players, Blockbuster, video cameras and a complete library of titles. All wiped out by a superior technology.

Now imagine the government had subsidized VHS to help it fend off insurgencies from new technologies. At best, many of us would still be stuck with an inferior technology. At worst, there would be no DVD. The government should let these things work themselves out in the marketplace; that's the true libertarian position.

All of this means that the government should let denser development happen. Stop the subsidies and tax policies -- including subsidies to new highways -- that encourage shifts to the suburbs. There are perfectly sound economic arguments for investing in rail lines in some situations; make those investments. Many home buyers, probably most, will continue to settle in the 'burbs. But we can't anticipate shifts in technology. Nor shifts in preferences or energy costs. We think we can. But we really, really can't. If the housing market would continue down today's path anyway, fine, but the government shouldn't lead it by the nose.

(Cross-posted at Austin Contrarian. Comment there.)

Yet another entry about high-speed rail

2009-09-05T15:31:18-05:00

The national bloggers continue to debate the feasibility of a high-speed rail link between Houston and Dallas. Up to bat: Megan McArdle.

I think reasonable people can disagree about the feasibility of HSR, especially since we're just trading back-of-the-envelope calculations of HSR's costs and benefits. Although I've argued HSR opponents underestimate the benefits to business-class passengers, in the interest of fair and balanced coverage, let me point to Richard Green's discussion of a more rigorous effort to tote up the costs and benefits. The authors of that particular study give HSR the thumbs down.

And while I'm in a contrary mood, let me suggest one other potential cost of a Houston-Dallas line.

Everyone agrees that in order for such a line to succeed, it must draw airline passengers to rail. Lots and lots of airline passengers. If the airlines lose, say, half of their Houston-to-Dallas traffic, they will have to cut the number of flights between Houston and Dallas. Obviously. Many consider that a feature, not a bug.

But what about the effect on other routes? American is able to offer service from Dallas to virtually every American city because of the huge number of passengers flowing into DFW every day. Cutting that inflow will reduce the economies of scale that support so many flights. Ditto with Southwest out of Hobby. Houston is a huge feeder for DFW, if the number of Bush-to-DFW flights is any indication. Similarly, Love is a huge feeder for Hobby, judging by the 24+ daily flights. Networks can be delicate things. Shrinking one route could sicken other segments.

HSR could very well make it easier to get from Houston to Dallas and vice versa. But it could very well make it harder to get anywhere else.

Cross-posted at Austin Contrarian; comment there.

The High Cost of Free Parking

2009-08-28T12:41:43-05:00

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

2009-08-12T20:46:58-05:00

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

2009-08-11T15:41:14-05:00

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

2009-08-09T16:08:20-05:00

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

2009-08-09T11:17:06-05:00

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

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.

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.

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.

Subsidized parking -- i.e., parking provided below cost -- distorts the market, encouraging an inefficient mix of driving and transit use.

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.)

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.

And there is evidence that other cities are getting by with less. The Manville and Shoup paper includes a very interesting chart (based on a 1999 study), which I've adapted below. It estimates the number of parking spots per job for various cities. It did not include Austin, but I borrowed the Austin data from Wilbur & Associates' 2000 parking study (pdf).

Obviously, many of these cities are oranges to Austin's apple. Foreign cities are not reliable benchmarks. Nor are old, dense American cities with extensive rail or subway networks. But there is no obvious reason why Houston, Los Angeles, Denver or Sacramento need less parking per worker. (And since this chart is based on 1999 data, neither Houston nor Denver's light rail accounts for the difference.)

The Wilbur & Associates study provides another comparative analysis. Again, Austin supplies significantly more parking per downtown worker than other cities (excluding San Antonio, which I suspect has such a high ratio because of job losses downtown).

Again, these are 2000 figures. Austin has added thousands of parking spots since then, excluding the many thousands added by the new condo developments.

The new condo dwellers have increased demand for parking downtown. But that demand has been absorbed (and then some) by on-site parking. I'm flummoxed why downtown needed a few thousand more commercial spaces.

I think this is evidence that Austin's downtown parking market is broken. The cause could be lender-required minimums, underpriced curb parking, the inefficient allocation of taxes to capital rather than land, or all of above and others. With the lull in commercial construction, we have a chance to diagnose the problem and develop a cure before the next wave.

(Cross-post and comments at Austin Contrarian.)