Monday, December 03, 2007

The last-mile problem with high-speed rail

In a Sunday Chronicle op-ed, Paul Mangelsdorf of Texas Rail Advocates laid out the case for high-speed rail in the Texas Triangle. I have to admit it sounds pretty cool, being whisked between cities at a comfortable 200-mph without most of the hassles of air travel or driving. But Texas has some key differences from Europe and Japan that make it a far riskier bet here - and we're talking about wagering many, many billions of dollars on this bet.

In telecom, they talk about the intractable "last mile" problem. It's relatively easy to run fiber optic trunk lines all over the country or even across a region, but connecting that "last mile" to every home is a very, very costly and difficult logistical problem - which is why we're still using our old cable and phone wire with relatively limited bandwidth. High-speed rail has the same issue: it can get you to the city, but what will get you that last 10-20 miles to your real final destination?

European cities were built at high-density in the walking age, and later had high-density transit systems grafted on top. Most of their key destinations - especially job centers - are usually in a relatively compact core. Using transit - or maybe taxis - for that last leg is not much of a problem.

Urban Texas was almost completely developed in the car age. Low density. Destinations spread out all over a large land area. Multiple job centers, many outside the core - not to mention long strings of offices and retail along freeways. It's simply not transit friendly. Yes, cities like Houston and Dallas are developing some rail transit, but they are very limited in the quantity of key destinations they can hit. The cities are just too distributed.

Even then, there's the cost of trying to bring that high-speed rail all the way into the core hub of the local rail network. I can't imagine what that will cost.

I know from a personal perspective, as cool as it would be, I would have trouble choosing high-speed rail to Austin for business and personal trips. I need to be able to get around Austin to multiple destinations . Taxis are unaffordable and a hassle, as is renting a car. Considering that an ultra-cheap one-way Greyhound bus ticket is at least $25, I can't imagine the rail ticket being less than 2-3x that. Pretty quickly, the math argues that I should just drive my own car there.

Lastly, not only does Europe have more population crammed into more cities that are closer together (perfect for rail), the airports in Europe are small and often overloaded - so trip shifting to rail becomes more attractive. All of the Texas Triangle cities have an overabundance of airport capacity for the foreseeable future.

The net is that Texas is not the ideal first candidate (also known as the "guinea pig") for high-speed rail in the U.S. Both California and the Boston-DC corridor have far stronger cases: more population, more density, more mature local transit, and more congested airports (and no, the Amtrak Acela does not count by this guy's definition of high-speed rail, averaging only 70-80 mph). The wise path for Texas in this case is to be a "fast follower" behind those two areas once they prove it can work, both technically and financially as well as ridership.

Labels: , , ,


At 7:42 PM, December 03, 2007, Blogger ian said...

Tory, I'm going to have to respectfully disagree (although, I'm up to hearing further arguments for the "wait and see" approach). At first glance, it seems to me that high-speed rail would be able to serve our spread-out, low-density Texas development at least as well as, and possibly better than, air travel does. I imagine riders would be able to park their car at the origin station and rent a car or take a taxi at the destination -- the same as they would if they were flying. Would rail really be any different?

In fact, I think it would be better. Because major rail stations would more than likely be located within inner city areas, riders would be much better served by transit options (or possibly even just walking!) than would flyers. Furthermore, the train could stop at several strategically placed stations on the way, presumably in the middle of suburban hubs where friends/family/business associates could easily pick riders up. That contrasts remarkably to most airports, which seem to be often located in remote locations, poorly served by transit, and easily accessible to only those people living nearby. Planes don't land at multiple airports along the way, after all!

What do you think?

At 7:49 PM, December 03, 2007, Blogger Michael said...

Well, I agree with your conclusion that Texas should be a fast follower of California and the Northeast. But I think California at least is already fairly far along in the process - they have a route selected, environmental studies underway, political support from the Governator on down, and a $10 billion bond to build the system from LA to San Fran will likely appear on the 2008 ballot. And the Acela, while it is not high-speed rail by European or Asian standards, already is a huge step up over anything we have here in Texas.

The last I heard, TxDot's "Texas Trans Corridor" does include high speed rail components, but it does not even go very close to our cities, because its purpose is not for the traveler but rather for freight / business.

As for "the last mile" problem, it seems to me the problem really has nothing to do with that. The problem is Southwest and the politicians here in Texas do not want this, because Southwest is still making some money off these routes, and flying them 10-30x a day depending on the cities. People who fly often face the same issue of renting a car when they get to their destination, but it does not stop them from flying. Furthermore, with Dallas and Houston's light rail systems, many business travelers working in CBD areas would not necessarily need anything other than trains / light-rail for transit (and maybe a cab ride or two).

The math is also starting to work out in favor of trains for the average person or business traveler. It costs 50.5 cents per mile to drive per the IRS 2008 rates, including depreciation on your vehicle. That works out to around $250 for your car trip to Dallas - likely coming out of either your pocket or your employer's pocket. If your train trip cost say $150 round-trip and got you back and forth in half the time, that's pretty close to your car trip including a rental car, parking, and / or a couple cab rides.

Plus, think of the possibilities - when this system is extended it will be possible to get from Houston to almost anywhere in the middle of the country (Chicago, Denver, New Orleans, Atlanta) in around 4-5 hours or less. When you take out the security checks / arriving 1-2 hours in advance for planes, and factor in that trains are something like 99.5% on time and are not affected by weather (except possibly in extreme circumstances), this makes the expected total travel time actually quite competitive with planes even for these mid-range routes.

At 6:52 AM, December 04, 2007, Blogger Tory Gattis said...

Yes, maybe a couple thousand people fly daily between the city pairs (mainly Dallas-Houston), but that's not nearly enough to justify high-speed rail. For comparison, the Main St. line moves 40,000+/day, and that was half a billion $ for only 7 miles, and the farebox revenue doesn't even cover operating expenses, much less capital costs (see here:

Now extrapolate that to *hundreds* of miles of precision track with far more expensive high-speed trains, but only carrying a few thousand people a day. The math gets ugly fast.

Flying uses existing capital assets we have to have anyway - airports - and people privately pay the complete cost of their trip, including covering the capital and operating costs of the planes.

Southwest's opposition is to gigantic government subsidies for a direct competitor - not unreasonable. A privately funded competitor would be fine, but the numbers simply aren't there to make that work.

At 8:41 AM, December 04, 2007, Anonymous Anonymous said...


While I agree that perhaps Texas is not the best place in the U.S. to start such service initially, your statement that people who fly "pay the complete cost of their trip" is simply untrue. If that were the case, there wouldn't be the hundred of billions of dollars spent each year on departments like the FAA, TSA, air traffic controllers, etc.

Airlines are one of the most highly subsidized industries in America. Personally, considering they provide such an essential service and given the inherent costs, I don't have much of a problem with some subsidization. But the same can also be said for rail. I would have no problem seeing some of those future subsidizes shifted to rail. If we keep relying primarily on air travel (our sky's are over crowded despite your statement to the contrary), we are going to have to upgrade our air traffic control system at the cost in the hundreds of billions. Perhaps we should think of an alternative, rather than just focusing exclusively on air travel going forward.

At 8:48 AM, December 04, 2007, Anonymous Anonymous said...

Don't airports have a "last mile" problem? Why is this different?

At 8:58 AM, December 04, 2007, Blogger ian said...

Nats - there's also the inherent subsidies related to air pollution and carbon emissions from planes that are far, far higher than those from trains. I bet if those subsidies could be adequately quantified, they'd be something in the hundreds of millions of dollars per year.

At 9:04 AM, December 04, 2007, Blogger Tory Gattis said...

My understanding is that passenger and airline fees completely cover the cost of air travel, inc. govt agencies. Big debate this year over how to allocate those fees between private and commercial flights to pay for the FAA.

We do have airport and air space congestion on the coasts, but not in Texas.

Airports do have the last mile problem. Part of what makes rail even remotely competitive with airports is *not* having that problem, which, unfortunately, it would still have in Texas. It just doesn't make sense to more than double your capital costs (airports + rail) while still having the same travel times and last-mile problems of flying.

At 9:07 AM, December 04, 2007, Blogger Tory Gattis said...

Even with carbon offsets, flying would still be more cost-effective. You can buy them now - it only adds a few dollars to the price of a ticket. And don't forget that most of the electricity to run that rail comes from coal-fired plants in Texas.

At 9:33 AM, December 04, 2007, Anonymous Anonymous said...

Well, I would be highly skeptical of any numbers showing that "passenger and airline fees completely cover the cost of air travel, inc. govt agencies." I think a simple investigation of the federal budget spent on the FAA and TSA would rebut that presumption. The airlines aren't refunding the federal government for such spending.

Not to mention the air traffic controllers, who are federal employees providing a service to the airline industry. Of course, this is also a subsidy. Interestingly enough, our rail network is controlled through dispatchers hired by the private railroads.

At 9:51 AM, December 04, 2007, Blogger Michael said...

Air traffic is heavily subsidized. And several of the airlines go through bankruptcy protection because they are not profitable.

Plus, from what I have heard, high speed rail is actually much more cost effective than comparable inner-city rail. I believe the routes in Europe/Asia are acutally profitable - they are geared towards business and tourism. Inner city rail is supposed to be more heavily subsidized and less profitable because its purpose is to move all people, even poor people going to their minimum-wage jobs, around in the most-efficient and lowest cost manner.

High speed rail to connect most American cities is a game-changer. It will be a revolutionary change, connecting our cities / businesses / tourism in a way that is simply not possible right now.

Tory, have you ever even ridden on high-speed rail? I find it helps to have experienced these different modes of travel before evaluating their impact / cost etc - and to have a better idea of their overall impact on the region.

Also, you are not doubling your capital costs by building rail. If Southwest were on board with this, they would get out of the Texas puddle-jumper flights for the most part, and rail would be allowed to be the main form of transit for this type of travel, where it naturally makes the most sense. Just think - high-speed rail gets you to Dallas in 1 hour and change, Austin in 40 minutes. Why do we need planes for this?

At 11:17 AM, December 04, 2007, Anonymous Anonymous said...

We need a good cost estimate for a high speed rail system in Texas. Until we have a good cost estimate, all this talk is basically pointless. There are many costly things we would like but we will never have (police, exemplary schools, lower college tuition, modernized freeways, rebuilt inner city, more parks, etc.)

I'll bet that the cost for high speed rail will be astronomical. At least $10 billion, maybe $20 billion or more.

But I'm just guessing, and so is everyone else until we have a good number. And the editorial writer should deal with financial issues and not just preach how great this would be.

At 11:26 AM, December 04, 2007, Anonymous Anonymous said...

A few observations:

"Europe" is not monolithic. In Germany, the train system can be charitably described as "Amtrak on a good day". It's a serviceable system, but it's not super highspeed, and Lufthansa makes a nice living handing intracountry travelers. If you are talking about true high speed rail ("bullet" trains), then you are really talking about Germany and Japan full stop. And that's not the full story. Japan - last time I was there in 2000- runs one Shinkansen line on Honshu. This line encompasses the major cities, but it's a narrow corridor hugging the pacific coast. If you are planning on taking a bullet train from Tokyo to sapporo, it's not going to happen. Similarly, France has a couple high speed lines, but the country is hardly blanketed. Both Japan and France have invested in enormous what I would call "interstate" highway projects in recent years, it's not the case that most travel within those nations is by train. The French and Japanese are too smart to run high speed train lines over commuter/freight tracks. It takes a particular kind of American bureaucracy to manage that feat.

Air travel doesn't require an enormous right of way acquisition the way rail travel does. Any discussion of the realtive cost and intrusiveness of air travel should take this point into account.

The notion that Acela works in the northeast is oversold. Acela still shares track with commuter rail and thus is a very low speed service over much of its route. It also travels over an ancient, windy right of way, complete with grade crossings in New England, further slowing matters. It would be great if we could get an arrow straight right of way, dedicated solely to high speed rail, and run the things at 150 mph. We can't, even if it were free for the asking, the NIMBYs in Westchester and CT aren't going to let the Feds mow down a 50 foot wide strip of New Rochelle or Greenwich to permit high speed rail, so we're stuck with trains travelling at a bit better than free-flowing highway speeds. All the enviro. obstructionist stuff that cyncically prevents highway projects is just as effective at preventing rail projects.

Finally, suburban office parkes are just as much a fact of life in areas where Aclea runs as they are in Texas. The last mile problem is very much an issue in the northeast as well. That said, current hassles involving security delays and flight delays in flying, and the location of Acela tracks through the center of downtowns (but just try replicating that in 2007 America anywhere) make Acela a time saver for many trips from central city to central city. You tell e if that's where the job growth is though.

What is the purpose of highspeed rail in Texas? Is it cheap mobility? Southwest airlines seems to provide that. Is it to have a cool-looking train system? That hardly seems worth billions. If you can find a private company to build and operate the system, with the gov't using emminent domain powers to help acquire the RoW, fine, but I doubt that's what proponents have in mind.

At 11:37 AM, December 04, 2007, Anonymous Anonymous said...

Are airlines "subsidized"? Could have fooled me. They are for-profit companies that pay the same taxes as any other corporation. They buy their equipment from for-profit companies. They pay special jet fuel taxes. My airline tickets are taxed into the stratosphere (I thought in order to fund the FAA, among other causes). The notion that airlines are "subsiized" because many have been through bankruptcy (repeatedly) is an odd argument. Why would a subsidized company be brought into bankruptcy? Doesn't sound like much of an effective subsidy to me. Besides, when a company goes through bankruptcy, claims against the company are reorganized (i.e., revalued downward). Those claims are private sector held claims, not government claims. You could argue that private investors and creditors are "subsidizing" airlines by making repeated losing investments, but even with that ultra-broad definition of subsidy, it's not a govenment subsidy that's at issue.

If you're talking about rail transit and subsidies and not mentioning Amtrak -- that seems quite the omission.

At 1:13 PM, December 04, 2007, Blogger Michael said...

Want to talk subsidies, and compare Amtrak and airlines? Granted, I am not a major Amtrak supporter - I think we need a much better system.


"Amtrak's entire budget accounts for less than one per cent of US Department of Transportation spending---$521 million vs. $33 billion for highways and $14 billion for air, not counting the post-Sep.11 bailout of $15 billion. -Source: US Department of Transportation"

As for Anon's comment that we need a true cost estimate for Texas, agreed. I just think when you and Tory say this, you don't necessarily really mean it. You mean "let's wait and do nothing until 2020", whereas I think we need to start evaluating this stuff more seriously now.

Really, what this type of program is going to take is a radical change from the feds on down. The "interstate highway system" for rail. And that probably is not going to happen anytime soon.

At 6:43 PM, December 04, 2007, Blogger Tory Gattis said...

Next time you buy a plane ticket, note all the special fees and taxes tacked on. The passenger facility charge covers the airport. The security fee covers the TSA. The FAA, including the controllers, is paid by the landing fees and fuel taxes paid by the airlines (not explicitly broken out on your ticket). Like I said, there was a large debate in the FAA reauthorization bill this year whether private jets should pay more of the FAA burden, since they take up just as much airspace to a controller as a 737. Right now the fees are heavily skewed towards large commercial jets.

The airlines have complained ever since 9/11 that TSA security should be paid for out of the federal budget (i.e. subsidized) instead of from fees on their plane tickets, since it's a national security issue. Of course, Congress has completely ignored that request.

> Tory, have you ever even ridden on high-speed rail?

Yes, the Eurostar from London to Paris. Beautiful ride. But I'd like to point out it connects the most dense urban conglomeration in Europe (it also goes to Brussels and connects to Amsterdam and the west German cities - at least 100m people in the zone), those cities have extensive local transit options for the last mile (not to mention dense compact cores), and it *still* bankrupted the company that built the Chunnel.

> you are not doubling your capital costs by building rail

I'm not talking about planes. I'm talking about airports. We have a giant existing capital asset for high speed travel in the Triangle - the airports. Adding rail creates massive redundant capital costs for providing essentially the same service. I understand that we wouldn't run some of those flights any more - but the airport costs stay. Many of the flights would actually stay, because they're connecting people to other flights, esp. AA at its DFW hub and CAL at its IAH hub, but even SWA connects a lot of people thru Love and Hobby.

At 7:06 PM, December 04, 2007, Anonymous Anonymous said...


The fees don't cover the costs of the TSA or other services done by the federal government on the airline industry's behalf.

This is for budget year 2003

TSA's budget was $4.8 billion. Fee's collected accounted for less than half of that ($2.2 billion). And that is only for the TSA alone.

Amtrak's numbers for 2005 are as follows:
Total Budget - $3 billion
Total Subsidy - $1.2 billion
At least Amtrak can fund more than half of its operations.

That is subsidy in my book.

At 7:24 PM, December 04, 2007, Blogger Tory Gattis said...

Yes, I believe they pushed through an increase in the fees since 2003 to cover the shortfall.

At 9:12 PM, December 04, 2007, Blogger Michael said...


I am not just talking about planes either. If you had a train system that serviced sub 500 mile routes, that would allow our existing airport infrastructure to be used as airlines focused on more profitable and more long range and international flights. Nobody is suggesting that we turn IAH into a garbage dump. But I think that using Hobby / IAH for shorter flights is not the optimal transit strategy from an engineering / efficiency standpoint (and not the best transportation strategy for the country either).

Duplication of capital costs is a nonsensical argument.

Railway investment would also reduce the need to invest as much in our interstate highways and provide things like mass-evacuation capacity in the event of hurricanes or terrorist attacks.

Apparently, Continental and American airlines already agree with me. Even Southwest may soon be on board if fuel prices keep rising.

And again, if you think the US airline industry is not heavily subsidized and protected from going out of business by the US government, I've got a bridge to sell you. If you can show me proof that the airline industry receives no federal subsidies, does not benefit through the TSA and tax-incentives, publicly financed infrastructure, etc - then I'll believe you. But I think saying that the airlines receive enormous help from the federal government and bailouts when required is about as well-known as saying that farmers receive subsidies not to grow crops. If you are saying it ain't so, you should show us some proof. Not from Randall O'Toole either ;) (unless he makes a great case).


At 9:24 PM, December 04, 2007, Blogger Michael said...

Or, you can just look at the numbers from the DOT:

Page 26:
$14 billion for the FAA for 2007

Show me how taxes on our tickets add up to $14 billion per year, and the problem is solved.

Surely some of that $14 billion is coming from ticket taxes, as you suggest.

At 9:47 PM, December 04, 2007, Blogger Tory Gattis said...

The airports have plenty of spare capacity. Cutting the Triangle flights adds no value. It would add no additional flights. In fact, it might cut some by losing connecting passengers. CAL and AA are on board because they'd love the govt to pay the cost of feeding them passengers (the rail plans connect to DFW and IAH).

> is not the optimal transit strategy from an engineering / efficiency standpoint

Actually, it is. Basic econ or business 101: if you build an expensive capital asset (like an airport), move as much volume through it as absolutely possible to minimize the unit costs. The key is having a high utilization rate.

> Duplication of capital costs is a nonsensical argument.

How about this analogy: Your house provides you with shelter. If you decide it would be nifty-cool to also have a high-rise condo with a view to stay in from time to time, and buy one, your costs have doubled. Yet in both cases, you're still 100% sheltered. Which is why the vast majority of people - especially financially constrained or prudent ones - only own one residence.

Both airports and rail get you to nearby cities at high speed. Why own - and pay for - both?

I don't know what to tell you about the airline subsidy topic. I follow the industry closely. They complain mightily about the tax burden, which can end up being up to 25% of the cost of the ticket. I don't have access to the figures. You're certainly welcome to believe what you wish. If there is a subsidy (and I'm nearly certain there's not), I can absolutely guarantee you it is trivial in comparison to the relative subsidy required to build high-speed rail.

At 11:19 PM, December 04, 2007, Blogger Michael said...


I don't think your analogy is very accurate. Owning a condo and house at the same time is an oversimplification of the argument. But, if you want to use that as the basis for comparison, then where I stand on the issue is more like:

-You already own a 1 bedroom condo (representing our existing highway AND air traffic infrastructure)
-Your wife is expecting and you've begun to notice strong putrid smells in the neighborhood and major repairs are needed to your place (our population boom and the costs of maintaining a polluting and inefficient existing infrastructure)

You can just say: "I will build another floor on top of my condo and keep fixing it up and buy an air purification system", which is a lot of money to keep sinking into your condo, or you could say "I will buy a better place, and either sell or rent out the condo" - thereby spending less money on "capital improvements" to the condo and getting yourself a much better place. This is, perhaps, Econ 102. But most people end up buying the house.

Again, rail replaces or at least significantly offsets the need for both future highway expansion and airport expansion - killing 2 birds with 1 stone. This is not the duplication of capital costs. This is the avoidance of continued capital investments in the wrong technologies and the wrong investments for our future (like the Texas Trans Corridor).

And even if rail did offset passenger demand at Hobby and IAH to some degree initially - I do not think the airports would go unused for long. You will just have more long-range flights - Dubai, Korea, Qatar. This is the future - and it is happening now. Likewise, I am not suggesting that we do not use and continue to invest in our existing road infrastructure. But expand it, by say building an interstate directly connecting Houston to Austin? That would be a prime example where I would favor building a rail line instead.

Finally, on the one hand you say high speed rail feeds customers to AA and Continental. On the other hand you say not a single flight will be added. Sounds like 2 different arguments to me.

One thing we can agree on: California will be a good test of the system in the US. We can talk about this again in a year after the voters approve their system and we know more.


At 11:38 PM, December 04, 2007, Blogger Michael said...

Also, to Anon:

>>If you are talking about true high speed rail ("bullet" trains), then you are really talking about Germany and Japan full stop.

Maybe I'm just talking about anything going about 100-150 mph or faster, but your statement seems to be pretty much untrue:

Planned / under construction High speed rail by country:

Existing high speed rail systems by country:

China, Turkey, Iran, Russia, Spain, Italy, Vietnam... heck even Croatia is building this stuff! Why not Texas?

At 3:29 PM, December 05, 2007, Anonymous Anonymous said...

That's a helpful link. However, (1) a lot of those systems are "proposed" or "planned" and (2) the entry indicates how fast the trains on those lines are capable of going, not how fast they actually go and over how long a stretch they operate at top rated speeeds. This matters a lot. Amtrak has Acela trains as we speak that are "capable" of running 150 mph from NY to Boston, but they only do so on a short strip of track near Westerly, R.I. (something like 10 miles or so). For the majority of the route, the trains are stuck at 79 mph (literally -- that's the speed limit) because they share tracks with commuter rail and operate on a congested and winding RoW outside the commuter zone. The trains actually are super slow when travelling through Manhattan and Queens. A plane, obviously, can fly as fast as its rating without regard to track maintenance or commuter traffic.

A lot of the lines in the entry cited travel at "up to" 100 or 130 mph. Great, but can you honestly imagine travelling from Dallas to St. Louis at that speed, assuming the route ran at that speed the whole way, which is dubious? It makes no sense as a plane substitute for a great deal of travel.

There is a notion out there that "bullet" trains are ubiquitous abroad, and that is not true. I think the entry linked to shows that they are becoming more extensive (note: on main trunk lines) in a few places, but for the most part "high speed" train travel is anything but. Ask yourself whether the German equivalent of the Austin/Dallas city pair would receive Deutsche Bahn's top-of-the-line equipment.

You can get from Austin to Dallas faster and cheaper on Southwest than a New Yorker can get to Washington by subsidized Acela. You should be happy about that, not resentful. I think high speed rail is "cool", and I'm happy that the Japanese taxpayers subsidized my fun rides on the Shinkansen for me when I visited that country -- I'd rather not return the favor though.

At 12:24 AM, December 06, 2007, Blogger Michael said...


>>Ask yourself whether the German equivalent of the Austin/Dallas city pair would receive Deutsche Bahn's top-of-the-line equipment.

Heck no. I can agree with that.

But Houston and Dallas are two of the wealthiest and largest cities in the Western world. They would rate as a pretty high priority no matter which country they were in - except maybe China where they would be considered smallish suburbs of Shanghai and Beijing.

>>You can get from Austin to Dallas faster and cheaper on Southwest than a New Yorker can get to Washington by subsidized Acela. You should be happy about that, not resentful.

As I've stated before, airlines / airports / the FAA are subsidized as well. I am a frequent flyer of Southwest - they are my favorite airline - don't get me wrong. I am not angry about the length of travel to get from Austin to Dallas by plane - at least the flight-time part of that. I also would not be angry by getting there by F-16, time-wise. But I also think it is not the appropriate technology given the trip distance. Kind of like taking a Southwest flight from Hobby to IAH - versus connecting the two via rail - if that is a trip that tens of thousands of people need to make on a daily basis. Rail simply is the best technology for sub 500 mile trips and potentially for sub 1000 mile trips depending on how much money you want to put into the line to get the most speed out of it. For places like Dallas, Austin, and San Antonio - with a maximum distance of 240 miles from Houston - rail is easily the best technology even at speeds of 150 mph.

Also, just FYI for anyone who cares, the numbers I have heard from California:
$82 billion - for extending the highway / airport system
$40 billion - to build the high speed rail instead to meet the same demand (which will ultimately go 220 mph from San Diego to San Francisco))

So, again, any talk of "subsidies" or "duplicate capital costs" is really disingenuous and speculative. Turns out in California's case, it is the complete opposite of the truth, as continued highway and airport investment / expansion that will be alleviated by rail would require over double the cost, over the same time period.

Of course, California's numbers will be different than ours - I am guessing ours might be more even at this point, but even Texas numbers are steadily going to drift towards rail making the most sense. I would feel much more comfortable if we could at least get a sense of what those numbers look like here in Texas, without biased politicians and corporations on either side dictating the outcome - and analyze the systems by their merits, not by what we have sunk money into in the past.

Let the technology that should win, win.


At 8:31 AM, December 06, 2007, Anonymous Anonymous said...

Whether rail can work for 500 mile trips is a matter that is dependent in part on the terrain. I think it would be virtually impossible to lay high speed rail track across Pennsylvania, New Hampshire or Colorado on any kind of reasonable budget. The terrain in Texas might be amenable to the type of RoW you need for rail, which would be a great cost-reducer. Not sure how the terrain works for an LA-SF route. California is quite mountainous between the coast and the I-5 corridor.

The subsidy point can't really be a source of dispute. There are real live highways in this country right now that generate tons of profits for public authorities and private companies. There are no profitable passenger rail lines in the U.S., and there won't ever be -- profitability is not even part of the business plan. Airports are money-makers. The Port Authority in NY-NJ makes money on airports and vehicle crossings and uses the money to subsidize passenger rail. The notion of "subsidy" for air travel really has to stretch the concept. American airports are rarely (ever?) privatized, but that's not the case worldwide. You can buy stock in Auckland, NZ's airport right now.

The 82 vs. 40 billion figure is instructive. Whoever put those numbers out there is effectively saying that Cali. can spend 82 billion to build adequate highway/airport capacity for the entire state and 1/2 of that comprehensive amount would fund rail on one and only one corridor. I'm actually pro-rail, pro-anything that improves mobility if it's a good deal -- but that sounds like a lousy deal to me.

At 8:33 AM, December 06, 2007, Anonymous Anonymous said...

California's airport system has been woefully inadequate for years. LAX has no room for expansion. San Diego, San Francisco, Oakland, and San Jose are being pushed to their maximums with land constraints. San Jose and Ontario Int'l have grown to pull demand from San Fran and Oakland and LAX, respectfully. Kind of like flying into Beaumont because IAH and HOB are overloaded.

California's highways has not kept with demand either. Many of the freeways within LA and the Bay Area still have the design standards CalTrans has instituted over 60-years ago. These are the standards Houston borrowed to design it's freeways around downtown and much of the inner loop. They were state of the art for many years.

Much of the increased cost for airport and highway expansion is related to the land and right of way acquisition costs. The same expansions for Houston, Dallas, Austin, and San Antonio Airports would cost much less. IAH's masterplan would allow it to be able to grow into one of highest capacity airports in the world. DFW has plenty of Terminal expansion it can add. Austin and San Antonio's airports are quite over sized for current capacity. Austin lucked out with a closed Airforce Base.

Highway right of way costs are also much cheaper. Even the cost overuns in the Katy Freeway would be minimal to a similar project in the urban California cities.

In Texas, the numbers would just be reversed for cost comparison!

This analysis has been performed several times in Texas by the Texas Transportation Institute.

At 11:39 AM, December 06, 2007, Blogger Michael said...


>>The 82 vs. 40 billion figure is instructive. Whoever put those numbers out there is effectively saying that Cali. can spend 82 billion to build adequate highway/airport capacity for the entire state and 1/2 of that comprehensive amount would fund rail on one and only one corridor.

Not true. This is a comparison of the route for rail versus highway. Not a comparison of the overall state highway budget versus one rail line.

>>The subsidy point can't really be a source of dispute. There are real live highways in this country right now that generate tons of profits for public authorities and private companies.

I think you could easily do high-speed rail in a profitable way as well. The concepts are the same as in most other forms of transportation - somebody needs to front huge capital costs and use eminent domain to build the system and acquire ROW. Then depending on tolls, etc, you could decide to operate for profit or at the public's expense, or partially subsidized. Remember, the overwhelming majority of highways are still operating at public expense.

KJB, do you have figures or links to the TTI's comparison of high-speed rail costs versus highway? I'm just interested to see the numbers / studies.

Again, current plans for Texas Trans Corridor do call for High Speed Rail - so obviously someone must have concluded this already makes sense even in Texas. I don't oppose the inclusion of the high speed rail component, but I do think the Texas Trans Corridor is a huge waste of money overall and is a seriously flawed concept - because too much money is spent on roadways, and not enough connectivity is provided to urban cores. And this is all still going to cost us somewhere in the neighborhood of $150-$200 billion to build.


At 12:19 PM, December 06, 2007, Anonymous Anonymous said...


Look at the link below.

It researches every document that TTI has performed regarding high speed rail.

Just in the first document I noticed it is estimated that high speed rail will cost $8.3 million dollars per mile in 1991 dollars. Downtown Houston to downtown Dallas is about 230 miles. This would be about $1.9 billion for one line. This value would be much greater at current values.

The TTC is not utilizing high speed rail. It was considered initially, but is not being implemented in corridors that are currently being designed and built that will eventually become part of the TTC. I worked on the design of the SH 130 (TTC corridor for the I-35 parallel road). We had to make allowances for dual rail tracks to eventually be placed in the median. High speed considerations were not taken into account for the grades of the railroad. The cost would have be astronomical. The TTC is primarily meant for freight movement. Passenger usage is possible by-product but not a consideration for initial use.

If you want a copy of the volumes of design documents and plan sets, the central TxDOT office in Austin would be more than happy to let you look at them since they are now public.

Also, you may have missed the point of the TTC. It is not to connect urban cores, but to do the best avoid them. It part of the concept that the urban cores are being inundated with through truck and rain freight traffic that a corridor to by-pass is warranted.

At 4:17 PM, December 06, 2007, Anonymous Anonymous said...

O.K. 30 seconds Googling ("FAA Budget 2007") confirms Tory's above assertion that (in the FAA's words), "The Airport and Airway Trust Fund, which is funded primarily by a set of excise taxes on commercial airline tickets, pays most of the FAA’s expenses."

I find it interesting that the same people who support/oppose local rail transit tend to also support/oppose high-speed intercity rail, even though they have virtually nothing to do with each other aside from the word "rail", and even though the debate is usually conducted in the absence of actual information about costs or benefits.

At 5:19 PM, December 06, 2007, Blogger Michael said...


"Most", as in 50.1% or greater. Nice work there. That also covers airport projects, corporate bailouts, etc, right? Didn't think so.

Also, as for supporting or not-supporting rail in the inter-city or region case - I find that interesting too, even though I guess I would be called a supporter. I prefer to consider myself fairly neutral, and the US fairly biased in its history towards favoring autos and air travel. Now our historical justifications for doing so are starting to make less sense - including sparse population, cheap natural resources, and examples from abroad that rail works just fine.

KJB - thanks for the link. I didn't see anything too interesting in a quick search but I will look more later.

As for the Texas Trans Corridor, I do think I get the main idea of it, as I mentioned in one of my previous posts - it is to help business, not residents. Sounds like a 150 billion dollar (or more) boondoggle to me. I predict that entire project will die or be modified beyond recognition before all is said and done.


At 5:36 PM, December 06, 2007, Anonymous Anonymous said...

I agree too about the TTC.

The only good aspect is that TxDOT realizes that they can't pay for it with taxes. They are building pieces through CDA's (comprehensive development agreements). As long as the agreement doesn't have a government bale out where taxes will pay for a bad deal, I'm ok with it.

The private companies financing the current pieces (Cintra - Spanish, Skansa - Dutch) and other projects using CDA's are in it to make money. If they don't see the tolls coming in, it isn't worth it. Luckily, no route using CDA's are critical routes. They are by-passes that trucking companies will pay for access.

The TTC freight rail concept works much the same way. The existing road built to accommodate rail will see that usage may need rail to prevent building more road. It will follow this patter. Build two lanes in each direction, then expand to three. If more capacity is needed, the rail portion will kick in. If they project demand is there, the rail may come in earlier. The beauty of the approach is that consideration for rail was in place so no right of way or serious grade adjustments need to be made. On top of it, it will not be subsidized by taxpayers.

It doesn't mean its perfect, but a unique solution that many parts of Europe are currently trying so they can expand and have a vast highway system like the US. All other state are looking at Texas right now to see how the first pieces come out. SH 130 will be the first leg to be built. It will start from I-10 around Seguin and parallel I-35 to the north side of Austin. North and south of Austin will have another toll segment to attached it to I-35.

At 10:25 AM, December 10, 2007, Anonymous Anonymous said...

Article in Popular Mechanics re: high-speed rail. I haven't read it, so I'm not commenting on the content, but many of you might find this interesting.

At 8:50 AM, December 11, 2007, Blogger engineering said...

Interesting discussion. A couple personal thoughts.
1. To connect Dallas and Houston with passenger rail is good but to detour via Waco and Temple to get to Houston does not make sense. Pluse the Dallas-Austin rail corridors are busy with freight already.
2. Need better definition of high speed train. 80 MPH is a confortable speed to fo from Houston to Dallas. Higher speeds are likely exponentially more in costs.
3. Mag lev technology has some environmental impacts and require (if not mistaken) 500 to 600 feet wide corridors.
4. High speed rail like the one in Shanghai in Texas? Why not, Houston to Mexico City.

At 12:11 AM, December 12, 2007, Blogger Michael said...


Thanks for the link from Popular Mechanics - that was an interesting article. For anyone else who wants to read it, the tiny url is:


At 11:38 AM, April 11, 2008, Anonymous Anonymous said...

Hi Tory,

I found your blog on the Texas High-Speed Rail project/proposal. You mentioned the Acela is not truly High-Speed and I am wondering where you get the speeds information from. Because, if you are right, then the Acela is indeed nothing but a regular train with a name of High-Speed attached to it. Other than that, I agree with your reasoning, but I don't think you are fully familiar with the California proposal and the Cal-reality, for it partially contains several of the Texas problems as well. I created a PPP on the flaws in Cal HSR and posted it at if you have remarks/suggestions to make, I'd appreciate that.


At 5:29 PM, May 08, 2008, Blogger Tory Gattis said...

Acela is high-speed relative to usual US train service, but not compared to high-speed networks in Europe and Japan.

I think you have an interesting analysis of the CA plan. I'm not really familiar with the situation out there and I don't know what the right answer is. But I do know that Texas should wait until CA makes it work before we consider it, given that CA has more people, more density, more local transit, and much more traffic and airport congestion.


Post a Comment

<< Home