A Rail Vision for America’s Future
posted in Transportation |Suppose Barack Obama is elected president and appoints someone like Portland Congressman Earl Blumenauer as Secretary of Transportation. Suppose further that California votes for high-speed rail. Then, even if some of the rail transit measures on the ballots in Kansas City, San Jose, Seattle, and Sonoma-Marin counties (have I missed any?) don’t pass, it is pretty clear there will be a strong push to build far more passenger rail in America.

How much rail is enough? How much will it cost? What good will it do? Let’s try to envision a rail future for America.
Start with intercity rail. The Federal Railroad Administration (FRA) has put together a plan for about 9,000 miles of high-speed rail. Under FRA’s plan, trains on most of the lines will move no faster than 110 miles per hour. But you know that, if Congress gives California billions of dollars to build a 220-mph network, then Florida, Illinois, and Texas will not be satisfied with trains that only go 110 mph.
Plus, three of America’s largest and fastest-growing urban areas — Phoenix, Denver, and Las Vegas — aren’t even on one of FRA’s proposed high-speed rail corridors. If federal funds are used for high-speed rail to such cities as Minneapolis, Orlando, and Seattle, you know that congressional delegations from Arizona, Colorado, and Nevada will demand their fair share of high-speed rail funds.

Adding lines from Albuquerque to Denver and Los Angeles to Phoenix via Las Vegas brings a national high-speed rail network closer to 10,500 miles. How much will it cost? Florida’s high-speed rail was supposed to cost about $25 million per mile. After taking into account increases in construction costs, as well as expenses needed to achieve higher speeds than the 125 mph proposed in Florida’s plan, and $35 million per mile sounds more reasonable.
High-speed rail in more mountainous California is currently estimated to cost around $50 million per mile. Not all of California’s proposed rail lines go through mountains, of course, so it is probably conservative to estimate that mountain segments will cost about $60 million per mile while flat segments are $35 million. If we assume that 8,000 miles of the national network are flat and 2,500 are mountainous, we get a total cost of $430 billion. That is probably very conservative.
How much urban rail do we need? It is hard to find a major U.S. city anywhere that is satisfied with its rail network; they all seem to want more. The New York urban area has more than 1,450 miles of rail transit, and New York City and New Jersey Transit are currently planning or building at least four more rail projects.

The big six transit markets — New York, Chicago, Philadelphia, Boston, Washington, and San Francisco — all have about 80 to 100 miles of rail lines per million residents. So let’s take 100 miles per million as a benchmark.
What kind of rail is needed by different cities? Urban areas of more than 3 million people tend to use heavy rail and commuter rail. Urban areas of 1 to 3 million tend to use light rail and commuter rail. A few urban areas with less than 1 million people — Austin, Honolulu, Madison, are also proposing various kinds of rail transit, and Salt Lake City (which was under 900,000 people in 2000) has light rail and is building commuter rail.
The big six transit markets have an average of about 20 miles of heavy rail per million residents, 10 miles of light rail, and close to 70 miles of commuter rail. Let’s set that as the standard for all urban areas of more than 3 million people (as of the 2000 census). Portland has 30 miles of light rail per million people, and is building or planning several new lines, so let’s take 50 miles of light rail and 50 miles of commuter rail as standard for areas of 1 to 3 million people.

For urban areas of 1/2 to 1 million, let’s set a standard of 40 miles each of light and commuter rail. We’ll use the same standard for areas of 250,000 to 500,000 (such as Madison), but only if they are on one of the high-speed rail networks.
Nationally, this comes to a total of about 1,500 miles of heavy rail; half of which already exists; 4,000 miles of light rail, 700 miles of which already exists; and 8,500 miles of commuter rail, 3,400 miles of which already exists.
To get some idea of costs, I looked at the FTA profiles of 2009 New Starts projects. Discarding a few outliers, heavy-rail lines are costing an average of around $180 million per mile; light rail $80 million per mile; and commuter rail (which usually uses existing tracks) around $11 million per mile. The result is a total cost of around $450 billion. Add at least 10 percent to account for population growth between 2000 and when the rail lines are built and you have a total cost of $500 billion.

Of course, the existing lines won’t last forever. As the Antiplanner has previously noted, New York City estimates it needs $30 billion to rehabilitate its subway system. As also previously noted, subway systems in Chicago, San Francisco, and Washington need about $39 billion. Each of these cities are projecting rehab costs of $100 million or more per mile, suggesting that rehab costs at least half as much as new construction. Add in Boston, Philadelphia, and other rail systems more than 30 years old and the total cost of bringing existing rail transit up to modern standards is well over $100 billion.
So, for a little more than $1 trillion, we can build a national high-speed rail network, rail transit networks in all major cities as well as smaller urban areas on the high-speed rail network, and rehabilitate older rail systems. If this system is built over a 20-year period, that amounts to $50 billion per year — about 5 times current expenditures on rail capital improvements. (Some of those current expenditures are going for rehabilitation, which the FTA counts as a capital improvement.) Once the system is built, we will have to keep spending an average of $500 billion on rehab every 30 years, or more than $17 billion per year.
Where would the money come from? Considering that the federal deficit is now nearly $10 trillion and growing rapidly, and considering that the credit crisis could add considerably to that deficit, Congress is not likely to fund transportation out of borrowing. The obvious source of funds is a gasoline tax, which proponents will probably bill as an anti-global warming carbon tax.
Each penny of federal gasoline tax raises about $1.4 billion per year, so Congress could pay for the entire rail system by increasing taxes by 36 cents per gallon — effectively tripling the current 18-cent tax. Or Congress could double the gas tax and pay for half the system, letting state and local governments find the money for the other half.
What will all this buy us aside from 23,000 miles of shiny new rails and trains? Europe has been spending a lot more than $50 billion a year on its rail systems, and as of 2000 rails carried only 7.4 percent of travel in the countries that were then members of the European Union. That’s a lot compared with the U.S.’s paltry 0.6 percent. But Europe’s rail share is declining in spite of all the spending, so there is no guarantee that the U.S. could increase its rail share anywhere near to Europe’s levels, even with the intrusive land-use regulation that typically accompanies rail construction.

In the United States, the urban area with the most rail transit is, of course, New York, where (as of 2005) slightly less than 10 percent of all travel goes by train. In no other urban area does rail carry more than 4.2 percent of travel. Even in Boston, which has the most rail miles per capita of any American urban area, it is just 3.1 percent.
Let’s be optimistic and say our trillion-dollar rail system will increase rail’s share of travel to 5 percent of the total. Americans travel between 5.0 and 5.5 trillion passenger miles per year, 5 percent of which is about 250-275 billion. Not all of this 5 percent would be drawn from the highways, but even if it were, highway driving normally grows by 5 percent every 2 to 3 years, so any congestion relief would be gone long before the 20-year build-out period.
Of course, highway driving declined slightly in 2007 and is expected to decline again in 2008. But gas prices grew by 30 percent in 2007 and another 35 percent in the first half of 2008. If prices stabilize — or at least stop growing so quickly — the growth in driving will begin again just as driving has been growing in Europe despite a long history of fuel costing $5 or more per gallon.
How does our rail vision compare with another transport megaproject, the Interstate Highway System? After adjusting for inflation, the Interstate Highway System cost less than half a trillion dollars). Yet it carries more than 1.0 trillion passenger miles (at 1.6 people per vehicle mile) each year, not to mention more than 0.5 trillion ton miles of freight.
Not only did the Interstate Highway System cost much less and move much more than our visionary rail network is likely to do, interstate highways have the virtue of being 100 percent paid for out of user fees. The rail system would require subsidies for pretty much all of the capital costs, most or all of the periodic rehabilitation costs, and at least some of the operating costs.
It is almost certainly this fact — that interstate highways were paid for out of user fees — that insured the system’s success. If people did not want to drive on interstate highways, they wouldn’t have bought the gasoline that paid for them. As it turns out, interstates make up just 2.5 percent of the lane miles of roadway in the U.S., yet they carry nearly 20 percent of all passenger travel and 12 percent of all freight.
There is no way — not even if gasoline reaches $10 per gallon — that our rail system, or any visionary rail proposal, will ever achieve this level of success. While rail, if it is powered by non-fossil-fuel electricity, could slightly reduce pollution and greenhouse gases, we could far more easily achieve the same results through plug-in hybrids and other high mpg or alternative-fueled cars.
Of course, the Interstate Highway System is pretty much complete, so the question is not whether rail is more efficient than it but whether rail is more efficient than other new transportation projects such as new roads, intelligent highways, and new auto technologies. Your answer may depend partly on what you think will happen to oil prices, global warming, or other factors.
For the Antiplanner, the paramount question is whether new transportation systems will pay for themselves or require huge subsidies. A system that pays for itself not only demonstrates its value but creates the right incentives for transportation users and managers. As much as I like trains, I see no evidence that any part of this rail vision can pay for itself, and so I conclude it is more of a nightmare than a dream.
If you support rail transit and/or high-speed rail, I hope you will post comments saying what kind of a national system you think we need, how much you think it will cost, and what kind of benefits you think it will produce. I would be especially interested if you think my proposal does not reflect the real desires of rail advocates and what you think their goals really are.




posted on September 29th, 2008 at 5:47 am
posted on September 29th, 2008 at 5:58 am
posted on September 29th, 2008 at 7:45 am
posted on September 29th, 2008 at 11:48 am
posted on September 29th, 2008 at 12:19 pm
posted on September 29th, 2008 at 2:12 pm
posted on September 29th, 2008 at 2:47 pm
posted on September 29th, 2008 at 2:48 pm
posted on September 29th, 2008 at 5:21 pm
posted on September 29th, 2008 at 6:16 pm
posted on September 29th, 2008 at 6:28 pm
posted on September 29th, 2008 at 6:52 pm
posted on September 29th, 2008 at 9:15 pm
posted on September 29th, 2008 at 9:52 pm
posted on September 30th, 2008 at 1:14 am
posted on September 30th, 2008 at 7:15 am
posted on September 30th, 2008 at 7:34 am
posted on September 30th, 2008 at 7:38 am
posted on September 30th, 2008 at 8:39 pm
posted on September 30th, 2008 at 9:06 pm
posted on October 1st, 2008 at 7:33 am
posted on October 1st, 2008 at 9:48 pm
posted on October 2nd, 2008 at 3:15 pm
posted on October 2nd, 2008 at 4:58 pm
posted on October 2nd, 2008 at 5:59 pm
posted on October 2nd, 2008 at 9:38 pm
posted on October 2nd, 2008 at 10:07 pm
posted on October 6th, 2008 at 4:01 am
posted on October 6th, 2008 at 8:01 am
posted on October 6th, 2008 at 1:47 pm
posted on October 6th, 2008 at 6:52 pm
posted on October 9th, 2008 at 9:18 pm
posted on October 11th, 2008 at 12:46 am
posted on October 26th, 2008 at 1:49 am