Urban Returns

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

From Streetsblog Capitol Hill:

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.

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

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

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

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.

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.)

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.)

TTI hyperbole

There is no question that congestion is a serious problem in this country -- and a crippling problem in some cities.  Nor is there any question that congestion is much worse than it was ten years ago, even if high gas prices and the recession have caused it to level off in some places.

But the Texas Transportation Institute is guilty of a bit of hyperbole when it announces that "[t]he overall cost [of congestion](based on wasted fuel and lost productivity) reached $87.2 billion in 2007 – more than $750 for every U.S. traveler."

TTI uses free-flow traffic as its baseline for measuring congestion.  In particular, it defines "Delay per Peak Traveler" to be "[t]he extra time spent traveling at congested speeds rather than free-flow speeds divided by the number of persons making a trip during the peak period." (Summary report p.1)

Honestly, free-flow speed is not the optimum baseline.  There is no realistic, hypothetical state of the world in which we would experience perfect, free-flow traffic everywhere.   It would not be feasible to build enough roads (or charge enough for them), particularly since free-flow speeds would entice more drivers onto the road.   So to imply that there is $87 billion of waste to be saved -- and I think TTI does imply this -- is simply wrong

Anthony Downs makes this point in Still Stuck in Traffic, pp. 23-24 (available on-line at Google Books).   He makes several other reasonable criticisms of TTI's methodology.   He nevertheless recognizes that  TTI's methodology and findings are useful, and he does not dispute (nor do I) that congestion is a serious problem that we can't simply build our way out of.

Punching above its class

The Texas Transportation Institute has issued its annual report on congestion in American cities.  As usual, Austin doesn't do well, punching well above its class.  Tthe average Austin traveler experienced an annual delay of 39 hours in 2007.    That's up from 32 hours in 1997, although no worse than 2006.   

Ryan Avent puts the Austin figures in perspective:

The average traveler in the New York metro area faces 44 hours wasted per year, for instance, while the average traveler in Los Angeles loses 70 hours per year to congestion, even though New York’s metropolitan population is much, much larger than LA’s.  More interesting still, Austin and Raleigh aren’t that far behind New York with 39 and 34 hours wasted annually, despite the fact that both metro areas have less than two million people while greater New York is home to 20 million people.

TTI also calculates the percent of peak period travel that is congested.   Austin again fares poorly -- its roads are congested for 70% of the peak travel period: 

By comparison, Dallas's roads are congested for only 66% of the peak travel period.  Houston is marginally worse at 73%.   Austin's peers include New York (69%), Detroit (71%) and Baltimore (69%).  

One has to wonder what damage Austin's congestion has wrought to downtown.   Businesses want to locate where their employees can get to them, and high peak-period congestion means that commuters must fight traffic to get to work most of the time.   I suspect Austin's downtown suffers more heavily from peak-period congestion than any other employment center in the area since MoPac and I-35 -- the principal commuter routes to downtown -- are the two most congested roads in the area.     Eventually we must recognize that our congestion will  stunt downtown's growth.   We wil lose the economies of scale and agglomeration benefits that a vibrant downtown would  yield.

You know, of course, that there is only one solution.