$100 Million a Mile

I don't really understand the various issues in this article on the next phase of Phoenix light rail expansion, but this certainly caught my eye:

It will add another $9 million to the $297 million project. But by
acting quickly to make these changes, there aren't expected to be
delays in rail construction. Work is scheduled to start in early 2009
and be completed by 2012.

Opposition to the rail plan arose last fall in the last half mile of
the 3.2-mile light rail line that extends from just south of Bethany
Home Road to Dunlap Ave.

Let's see -- $306 million divided by 3.2 miles is very close to $100 million a mile, and that is even before the inevitable cost overruns cut in (as a rule of thumb, I tend to double estimates of light rail construction costs to estimate the actual final total, and even then I am often low).   It also does not include inevitable operating losses.

Nearly a third of a billion dollars to run a rail line a distance most people could walk in 45 minutes.  For three freaking miles.  As a comparison, three buses could provide service on this same route running at 5 minute intervals for perhaps 1% of this capital cost and a substantially lower operating cost.  And better service, since the frequency would be 3 times higher.  Absolutely absurd. 

More on Phoenix light rail here, and more on light rail in general here.

Postscript: Some of you may be familiar with my light rail bet.  I often bet that a light rail line will cost more to build than it would have cost to buy every  regular daily rider a Prius, and more to operate in a year than it would require to gas up all of these Prius's for a year.  For reference, with a $22,500 cost for a Prius and $306 million (and counting) capital cost, that is enough to buy 13,600 Prius's.  Anyone want to bet that the number of incremental users attracted to the line by this 3 mile extension don't exceed 13,600?

Update:  TJIC does the math -- $1500 per inch!  Fixed link, thanks to commenters.

9 Comments

  1. Porkopolis:

    I think your link for the Arizona Republic article is not working. I think you were trying to point to this:

    http://www.azcentral.com/news/articles/2008/07/02/20080702phx-lightrail0705.html

  2. K:

    I live outside Phoenix. To the political establishment of AZ the rest of the state exists for the purpose of sending money into a few square miles of central Phoenix. That system isn't perfect but they try.

    Light rail is just another fringe benefit for those who work in that central area. They are - surprise - mostly government employees or employees of government funded organizations such as state universities and the new medical school.

    The AZ Republic and the local TV function as a public relations department for anything which will expand government.

    After spending enough they will have the trains running on time.

  3. K:

    I live outside Phoenix. To the political establishment of AZ the rest of the state exists for the purpose of sending money into a few square miles of central Phoenix. That system isn't perfect but they try.

    Light rail is just another fringe benefit for those who work in that central area. They are - surprise - mostly government employees or employees of government funded organizations such as state universities and the new medical school.

    The AZ Republic and the local TV function as a public relations department for anything which will expand government.

    After spending enough they will have the trains running on time.

  4. Porkopolis:

    This is interesting...Tucson is able to build light-rail at $20 to $40 million per mile:

    "...cost: Basic at-grade light rail construction is currently running approximately $20-40 million per mile (including all capital costs, maintenance facilities, and rail cars), while new highway construction has typically cost $40 million per mile or much more..."

  5. Dan:

    Warren, you're missing the positive externalities of light rail. How do you intend to account for the warm fuzzy feeling that it'll give to Arcadia neighborhood liberals who will never ride the thing? And truly, isn't that worth half a billion dollars?

  6. vanderleun:

    Light rail is municipal heroin.

    Once the needle goes in, it never comes out.

  7. John Moore:

    An urban planner working for Phoenix once explained this to me...

    Downtown exists so that the financial sector and legal sector can be close together and close to the courts and government offices [yeah, this was in the internet age]. That downtown is critical to the well being of the city, and light rail will allow business to locate downtown and grow because their employees will have transportation. That will create an environment where local service businesses will flourish, and that's good.

    Honest... that's what they're spending all this money for (plus whatever graft they can get).

    Basically, someone who should have stayed in Manhattan wants to transplant its good features (for big city people) to downtown Phoenix. And they'll do it by building a rail system that goes through a tiny fraction of the city, which magically employees will appear at to ride to work.

    Light rail is another item of the liberal religious faith. It is sorta tied to saving the environment (higher density), building up metrosexual heaven, central government planning (we know what's best for you) and good citizenship (i.e. riding to work on crime-ridden public transport) - or some such rot.

    But hey, these same geniuses taxed us to build a stadium downtown when they should have been building freeways between the suburbs.. hmm... which is what the light rail money should have gone to.

  8. Allen:

    The Tuscon light rail site is at best ignorant if not knowingly lieing.

    At-grade LRT does nothing to unite a neighborhood. Track crossing are limited. Where you can cross, traffic frequently backs up. The line is at-grade, cars have to stop every 5-15 minutes to let the train go through. And those cars are forced into choke points due to the limited number of grade crossings.

    The tracks do divide a neighborhood. People by their nature aren't big on cutting across them. They'll follow the sidewalks. And those will cross at the few limited grade crossing. Where people don't follow the sidewalk, they will have to deal with crossing the tracks with trains coming by every few minutes. More so, what neighborhood would allow the tracks to not have some sort of fencing to make sure their children don't go there to play?

    Light rail generates a tremendous amount of pollution. The difference is instead of being local, it's moved off to where the power is generated or where the steel and concrete are manufactured.

    "Study after study has shown that building freeways to relieve congestion actually only creates the need for more freeways to reduce congestion;"
    There are no studies that show building more lanes CAUSES more driving. Not one. Think about it, if the existence of that concrete could actually cause people to drive more, places like Buffalo and Pittsburgh would have traffic problems far beyond what they have.

    "26,300 cars per day currently drive on 6th Street between Euclid and Campbell; light rail can move more people along the same segment in less than one hour. To put it another way, two Light Rail tracks carry the same capacity as 16 lanes of freeway."

    What this doesn't address is where those cars are coming and going from. Chances are a large share if not 2/3 or more of them do not originate nor have a destination on that corridor. They're simply passing through. LRT on that route would not adequately address those needs.

    More so, as pointed out above and below with numbers, 2 light rail tracks will not carry any more than than 2 lanes of freeway. The capacity the refer to for LRT doesn't exist. Stations have capacity limitations. Trains have spacing and length limitations. And, as just pointed out, people will only take the train if it meets their needs. If you're just passing through, you're not going to stop, take the train, and have a 2nd car on the other end waiting to continue your journey. 16 lanes of freeway? The T-Rex project I mentioned, has 10 lanes of freeway in it's widest areas. It handles 280,000 cars today and there's room for growth. That's @450,000 rides. And that's just 10 lanes. Double track LRT can not handle 450,000 riders per day. Even using their numbers, 200 people per 2-car train, assuming 4 car trains, and trains running every 6 minutes for 24 hours, they could only handle 144,000 riders. 10 lanes of freeway has 4-5 times that CAPACITY. Let alone what people are actually using. As we know, there are only a couple hours in the morning and evening for high demand.

    Their effeciecy comparison doesn't make sense. A single lane of freeway can carry, IIRC, 2500 cars / hour. USDot's vehicle occupation rate is not 1.1 but 1.6. LRT is limited to the size of the station and spacing of the trains. After all, you can't simply have trains sitting still on the tracks. What it all translates to are things like the recent T-Rex project in Denver. The freeway portion, I-25, carries 280,000 vehicles per day. That means it has @ 450,000 "riders" per day. The parrell LRT double track (about the width of 2 freeway lanes) carries @ 38,000 riders per day.

    The cars off the road claim is at best poor polling. Studies have shown that 2/3 to 3/4th of light rail line riders were already taking the bus. Fancy that, people who were already pre-disposed to taking transit will keep taking it. So, for that double tracked line I just mentioned with 38,000 riders per day, the one that takes up the space of 2 freeway lanes, you have about 4,000 less cars on the road (keep in mind a "rider" is a person on one trip; an individual that makes 2 trips in a day on LRT is 2 riders). Oh, that's less than the capacity of the freeway lanes that could be in that space! More so, they don't remind you of LRT's other dirty secret. The majority of riders drive to the station. Talk to a traffic engineer and you'll learn it's keeping that local street traffic moving that gets tricky, not transporting people for long distances. So what happens is most people drive to the station while the transit authority takes those buses that were running the route and sends them out on local routes to connect to the station... only they run nearly empty. Even if every LRT rider was new to transit, they'd still be driving their car. Just a few less cars on the freeway concentrated into local traffic spots. That is, it shifts the problem to the local citys which rarely get adequate funding to address them.

    The cost projections are farcical. Denver's Fastrack's West Corridor is, after $275 million in cost cuts, going to still cost over $52 million per mile. That is after a variety of cost cutting measures that include single-tracking the last few miles of the line, reducing station size from 4 to 3 cars, building it to withstand 5 year storms instead of 100 year storms, etc. Granted the entire line is not being built at grade. There will be a couple bridges over very major streets. But what so called at-grade line would at least not include a couple of those? No sane traffic engineer would want a 4 - 6 lane boulevard or highway to have a grade cross that has trains coming through every 5 minutes. It'd be cheaper to not build even a couple of those bridges but it wouldn't be practical.

    But really that link, http://www.tucsonlightrail.org/vsfreeways.htm , as a whole points out the lack of meaningful comparisons. They drudge up the worst case scenarios for cost and apply them to the entire freeway. They also avoid taking about actual ridership and instead dwell on the issue of capacity for the size of the footprint. But even when they talk about capacity they don't take into account how frequently trains can run nor the physical limitations of the size of the stations. Theoretical capacity is interesting but what matters for actual projects is how they will actually be used.

  9. linearthinker:

    OT. HOV lanes in AZ (Diamond Lanes elsewhere):
    A huge waste of R/W acquisition and construction funds.
    Every time I've traversed Sacramento I marvel at the stupidity. Off peak and peak hour traffic...nearly empty Diamond Lanes. I've seen it while traversing Phoenix also, till I discovered there's a way to completely bypass Phoenix if you're map literate and don't mind a few extra miles. I love the grade separation structures dedicated to near empty HOV/Diamond lanes.
    Consider also the hazards introduced by HOV lane users having to negotiate crossing 4 or 5 lanes of rush hour traffic to enter/exit the freeway without missing their exits. Progressives smugly discount reality. Gaia smiles.