Tuesday, May 18, 2010

The Case(s) against Rail Transit (and the financial collapses in Dallas and Austin)

I've been building up a list of recent stories on national problems with rail transit projects.  The most comprehensive is Randal O'Toole's new report at Cato on "Defining Success - The Case against Rail Transit".  It is packed with disturbing statistics.  Here is the Executive Summary:
Over the past four decades, American cities have spent close to $100 billion constructing rail transit systems, and many billions more operating those systems. The agencies that spend taxpayer dollars building these lines almost invariably call them successful even when they go an average of 40 percent over budget and, in many cases, carry an insignificant number of riders. The people who rarely or never ride these lines but still have to pay for them should ask, “How do you define success?”
This Policy Analysis uses the latest government data on scores of rail transit systems to evaluate the systems’ value and usefulness to the public using six different tests:
  • Profitability: Do rail fares cover operating costs?
  • Ridership: Do new rail lines significantly increase transit ridership?
  • Cost-Effectiveness: Are new rail lines less expensive to operate than buses providing service at similar frequencies and speeds?
  • The “Cable Car” Test: Do rail lines perform as well as or better than cable cars, the oldest and most expensive form of mechanized land-based transportation?
  • The Economic Development Test: Do new rail lines truly stimulate economic development?
  • The Transportation Network Test: Do rail lines add to or place stresses upon existing transportation networks?
No system passes all of these tests, and in fact few of them pass any of the tests at all.
His short analysis of Houston is on p.25.  He also calculates that Metro loses $3.27 for each passenger trip on the light rail, which is actually not bad compared to most of the lines in the list (and in line with the loss subsidy for bus trips).  But I would expect the loss to be much, much worse on most of the new lines.

His conclusion:
Since around 1970, the modern rail transit boom has led American cities to spend close to $100 billion building, and billions more operating, new rail transit lines. This analysis indicates that these new lines almost always waste taxpayer dollars. Instead of providing cost-effective transportation, rail transit mainly transfers wealth from taxpayers to rail contractors, downtown property owners, and a few transit riders who prefer trains to buses.
Most of the few rail regions that enjoyed increases in per-capita ridership or transit’s share of commuting supplemented rail construction with strict land-use rules that reduce housing affordability; transportation plans that deliberately increase congestion to discourage driving; and subsidies to high-density transit-oriented developments along the rail lines. The costs of these policies are high and benefits negligible.
By almost any objective criteria — profitability, ridership, cost-efficiency, the Cable-Car Test, economic development, and the effect of rail transit on a region’s transportation system as a whole — few American rail transit systems make sense. Congress should correct the perverse incentives that encourage transit agencies to choose high-cost solutions to transit problems. Transit agencies should stop building rail transit. With the possible exception of a few subway and commuter-rail lines in New York and one or two other major cities, agencies should make plans to shut down existing systems when they are worn out and would otherwise require expensive rehabilitation. Those exceptions should be maintained only if they can be locally funded, preferably out of user fees and not general taxes.
Moving on to stories on specific transit agencies and their rail plans:
(Hat tips to kjb, Brandon, and BlogHouston for story links.)

Can you see how it might be a blessing in disguise if we lose the $900 million in federal subsidies for the N and SE LRT lines and instead did a re-think of the Metro Solutions plan before we join the same headlines above?

To wrap it up, a quote and some thoughts:
“Why do politicians love trains? Because they can tell where the tracks go. They know where everybody’s going. It’s all about control. It is all about power. Politics itself is nothing but an attempt to achieve power and prestige without merit. That is the definition of politics. Politicians hate cars. They have always hated cars, because cars make people free. Not only free in the sense that they can go anywhere they want, which bugs politicians in the first place, but they can move out of the political district that the politician represents.”
--P. J. O’Rourke, “Driven Crazy,” Reason, November 2009, p. 15.
Not necessarily completely accurate, but (darkly) funny.  I think it stems more from a fundamental disconnect.  Voters aren't happy with their commutes and they've experienced trains in older cities or other countries that work, and wonder why they can't have the same thing - even though newer cities have fundamentally changed around the car in the last 60 years, and the economics of rail has deteriorated to the point of infeasibility (outrageous costs, few riders).  These arguments are too complex to easily convey to voters, so politicians do the simple thing: promise rail, and either study it indefinitely because it can't be made to work (or  "we're waiting on federal funds"), or actually build something that we're ultimately disappointed with, either because it was done on the cheap and doesn't connect where people really want to go, or it has massive cost overruns that bankrupt the local transit agency (or, in some cases, both).  This is also why you don't see almost any agency building any rail without massive federal subsidies, because those subsidies are the only thing that make it remotely feasible.  And they still run into cost problems, even with the subsidies, and ridership is almost always disappointing.  Yet the rail momentum seems unstoppable, certainly by anything as flimsy as logic.  Sigh.

Labels: , , ,


At 8:46 PM, May 18, 2010, Anonymous Steve said...

Thanks Tory. While I sometimes feel Mr. O'Toole overstates things a bit, (though I haven't read the piece that is the subject of your post), I can only hope that he helps lead Houston to a more sober assessment of what is realistic transit-wise.

I think we do need improved transit, but given the physical nature of our city, doing most kinds of transit are difficult. Why are we not exploring all sorts of different ways to apply bus service? METRO promised a lot several years ago regarding buses - a revamped route network, new "Signature" service, etc. The only improvement that's happened is the Bellaire QuickLine, a positive step but pretty weak in terms of actual application (limited operating hours, route terminus too far east). Where's the other improvements? Houston's urban core densification provides lots of opportunities for improved buses - why do we hear nothing on this? Instead, all effort (and $) seems focused on light rail - including some pretty low-density corridors (even if they have a relatively high presence of transit-dependent travelers).

Not to mention so many ways to make the city more walkable as it densifies - badly needed improvements that are generally lower cost than whole new transit systems - and isn't that a pretty basic requirement for transit success?

At 10:03 PM, May 18, 2010, Blogger Tory Gattis said...

Total agreement. There are so many innovative things we could do with buses if the focus on LRT wasn't all-consuming.

At 11:12 PM, May 18, 2010, Anonymous Martin said...

So, following your general logic, would you say that auto-dominated transportation is also a failure because of TXDOT's current fiscal situation?

At 12:30 AM, May 19, 2010, Anonymous Anonymous said...

I agree that LRT has been focused on too much lately in transportation planning and discussion and feel that some cities have overexpanded rail too quickly. However, this logic put forth is simply incorrect. We need options in our transportation and having LRT connecting our major employment centers together is a no brainer because buses simply will not do. Having the SE and East line is a crappy use of LRT and our resources should be focused else where, like the Uptown and University lines and improving connecting bus service.

And well said Martin.

At 8:38 AM, May 19, 2010, Blogger Tory Gattis said...

TXDoT has not had an inflation adjustment in the gas tax for 20 years, so it has had steadily diminishing resources, which, for the most part, it has invested as best it can in the highest value projects (with some exceptions for the political power of rural areas). TXDoT know its mission and has pretty much stuck to it.

On the other hand, transit agencies have consistently underestimated costs, overestimated revenues and ridership, and generally mis-invested limited resources in projects with low value, pretty much no matter how you want to measure 'value'. They have abandoned the mission of mobility for the glamor of rail.

At 10:59 AM, May 19, 2010, Anonymous Martin said...

The current financial problems and issues with transit agencies (especially here in Texas) are based on falling sales tax revenues which are a result of the recession and financial crisis. The fiscal crisis at TXDOT is based on a longer-term problem; people don't want to pay higher taxes on gasoline and they also don't want toll roads. Yet at the same time people have voted to increase their sales tax to pay for transit.

The public generally supports better transit alternatives. In many cases that includes rail. Ridership has been increasing on most systems. People also want more walkable and safer communities. You can look at poll results, real estate demand and prices and the results of referendums to see that is the case.

Have there been circumstances where mismanagement has occurred with transit agencies? Of course. But to try and claim that the same hasn't happened with TXDOT is laughable to the extreme.

At 12:45 PM, May 19, 2010, Anonymous Neal Meyer said...


It was during the 1930's, when the United States and the rest of the world were mired in the depths of the Depression, that transit companies - which were at the time privately owned and operated - decided to abandon their rail and street car operations and go with bus service. Rail transit survived in only a handful of older American cities after 1940, as private companies realized that they could not maintain the capital investment needed to sustain rail.

In many ways, the overall economic outlook regarding transit is similar, save for the fact that transit companies are now local government agencies and that changes everything.

TX-DOT's annual budget is roughly $9 billion per year for the entire state, but even though that figure pales when compared to how much money the State spends on government schools or Medicaid, it's still $9 billion too much for some of my hardcore libertarian friends. Yet, whenever I ask them if they'd prefer tolls to pay for roads, support more localized government control and funding, abolish TX-DOT, or to privatize roads, they still flip out and still say no - they wouldn't support any of those. Ergo, they won't ever give out any answers and effectively they don't have any answers.

Screw 'em.

At 1:21 PM, May 19, 2010, Blogger Jardinero1 said...

The problem with the analysis is that the questions are framed where a system, rail transit, is expected to perform certain functions and pay for itself at the same time. But can it do that if it is competing with a commons i.e. free roads and highways? There is never an expectation on roadways or road construction that another lane of highway or a new street laid out in the suburbs is going to be cost effective.

As long as drivers are allowed unlimited, free use of roadways, rail transit or any other mass transit can never be cost effective.

On the other hand, I would submit that many forms of mass transit such as rail, gondolas, cable cars, and buses would be not just cost effective but profitable if individual drivers had to pay by the mile for their use of the roadway and the per mile cost was as much as the market would bare to keep the road full but flowing.

The gas tax and local property tax don't fit the definition of user fees. True user fees tell the owner operator of the roadway what sections of the roadways users really want to drive. User fees can rise on roadways that users demand more of. With price rises, the market will dictate fewer individual users will use the roadway and more mass transit will use the roadway.

Gas and property tax do not convey that information to the owner operator of the roadway or allow rapid adjustments in the use of the roadway i.e. a transition to fewer individual users and more mass transit. If I am paying gas tax and City of Houston property tax, another entity downtown or in Austin or in DC is deciding where to spend the funds based on a political process not a market process.

At 3:26 PM, May 19, 2010, Blogger Tory Gattis said...

I agree that gas taxes are not full congestion pricing, which we should definitely be going to. But a dollar spent on roads definitely facilitates many more trips than a dollar spent on transit. The returns are not even close.

If you think about it, it's sort of obvious: build a lane-mile of simple pavement, and thousands of people will drive on it with vehicles and energy they pay for out of their own pocket. With transit, the vehicle, network, and energy are all paid for (and fares barely cover 20% of the cost), so it definitely costs more from the govt perspective to move someone by transit. Govt has to pick up more of the total tab with transit.

At 11:21 PM, May 19, 2010, Blogger Alon Levy said...

Tory, the NTD says that METRORail costs $1.34 to operate per passenger. Not $3.27. You don't need thinktank fellows to calculate those numbers; they're all publicly available.

At least O'Toole isn't saying something as stupid as "politicians have always hated cars." So Eisenhower wasn't a politician? The people who put Robert Moses in charge weren't politicians? Hoover wasn't a politician? What the hell?

At 7:15 AM, May 20, 2010, Blogger Jardinero1 said...

I think I agree with you Tory. Roads are functionally much more useful. Through user fees or congestion pricing market forces would induce mass transit on the same roads, be it car pooling or buses. Government run mass transit would be unnecessary.

At 7:52 AM, May 20, 2010, Blogger Tory Gattis said...

The $3.27 is a loss per trip, and is probably based on total costs, including construction amortized over some number of years.

Looking at your data, it says it costs $1.34 in operating expense per trip, and I think the revenue is $1, so that implies not much of an (operating) loss. But higher up it says light rail costs $15.9m to operate and only brings in $5.3m in revenue. Something doesn't match up.

At 7:54 AM, May 20, 2010, Blogger Tory Gattis said...

Oh, and just to clarify: I think the Main St. line has been a reasonable investment. But it connected many major destinations in a relatively short corridor (something rail is perfect for). Future lines won't have anywhere near the same performance. The Universities line will be the best, and I doubt it will even come close to matching the Main St. line.

At 9:51 PM, May 20, 2010, Blogger Alon Levy said...

Most riders probably don't pay the full fare. They either transfer for free, or use reduced fare or unlimited cards.

The Uptown Line should perform well - not just the Main Street and Universities Lines. Arguably it would be better to have University turn north to serve Uptown in the first stage instead of Hillcroft; Uptown's big enough of an employment and retail center to justify it.

At 7:49 AM, May 21, 2010, Blogger Tory Gattis said...

I've thought something similar might make sense: have the U-line have a spur up to the Galleria, then backtrack back south to Westpark and out to Hillcroft. The extra stop and distance would only add a few minutes to the route, and it would add a major destination.

At 3:39 PM, May 21, 2010, Blogger Alon Levy said...

Ugh... detours generally don't work well; rail operators use them only when there's no option. Even if they don't add too much time, they annoy passengers, much like persistent slow zones and long wait times.

If there's a University-Uptown line, there's no reason to go to Hillcroft. There's no anchor there. The retail near the Southwest Freeway would be one stop further out of Hillcroft.

At 3:58 PM, May 21, 2010, Blogger Tory Gattis said...

The Hillcroft transit center is a major transit hub, the Gulfton area south of Westpark has the highest density of transit riders in the city (check out the satellite maps of the low-income apartment complexes), and there is a significant retail hub at Harwin and Hillcroft, right next to the transit center. The U-line would be giving up a lot of riders ending at the Galleria instead.

At 12:36 AM, May 22, 2010, Blogger Alon Levy said...

Ah. Fair enough...

At 10:12 AM, May 24, 2010, Blogger M1EK said...

I can't believe somebody from the metro area that was the case study for how poorly the gas tax operates (what did the study say, 16 cents on the dollar now?) would throw back the "let's build more highways" kool-aid.


Austin's disaster is not because rail sucks; it's because BAD rail sucks. Houston's problems are far from disaster status and are largely an issue of being over-ambitious; the plans themselves are fine as far as I can see.


Post a Comment

<< Home