Wednesday, March 2, 2011

Krugman gets it wrong about High-Speed Rail

Paul Krugman is Nobel Prize winning economist and an economics journalist for the Opinion Pages of the New York Times.  How dare I contradict what he has to say about anything?  Well, here goes.

Mr. Krugman has been a high-speed rail advocate for some time and for several reasons.  Previously, his discussions stressed the economic benefits and the jobs that would be derived from a national high-speed rail program.  That, I suggest, is the primary reason for the Democratic Party support for this Obama program.  They see it as a stimulus program, not a transportation solution.

But now, Mr. Krugman states that it is faster than flying to take the high-speed train between Boston and Washington, as well as between San Francisco and Los Angeles. He's no longer talking about his field of economics, but about comparing transit modalities, and that's the Department in another building on campus. 

In an un-numbered, abstract diagram, below, he shows how the two time/distance curves of plane and train overlap so that the shorter the trip, the faster the train travel time.  In the abstract, all other factors being exactly the same, that may very well be true within the parameters Krugman specifies. But, we are not talking in the abstract.  We are talking about real places, where real money is going to be spent and wasted.  Lots of it. 

Am I about to quibble with Mr. Krugman here?  I don't think so.  Just like Mr. Krugman, the CHSRA justifies the HSR project on the grounds that it will be faster to go from San Francisco to Los Angeles by their train than by flying.  This is a theoretical claim, as is their forecast that they will make the train trip in 2 hours and 40 minutes.  

Well, even if it's true theoretically, it's wrong in the on-the-ground application.  To begin with, although important, top speeds are not the only crucial variable; average total trip speeds also are.  So, the total trip times between LA to SF air terminals and rail terminals are more meaningful comparisons.  But even more meaningful are the door-to-door times.  

There should be no disagreement that if both a HS train and a plane leave their respective terminals simultaneously planes will arrive sooner. Planes fly three times as fast as the fastest trains. However, regardless of their top speed capability, the high-speed train takes many diversions from a straight line, unlike planes.   High-speed trains cannot go top speed leaving or entering population areas; the larger the urban area, the slower the average speed.  Furthermore, train acceleration is far slower than that of planes. In other words, over the ground, even the fastest trains are still much slower than commercial airlines.

The CHSRA promise of a 2 hour and 40 minute train trip has yet to be proven.  I strongly suspect it's an empty claim.  The route wanders down the Bay Area Peninsula from near the Coast to inland through the Central Valley, only to snake back to the Coast upon approaching the LA Basin.  It must slow down as it passes through many of the towns along its route. 

If Mr. Krugman believes that in 10 years, we will be able to waltz on board our high-speed trains, stow our luggage and take our seats, he's just not reading the newspapers.  There will be security, luggage X-Rays, searches for weapons, etc.  Our Department of Homeland Security will see to this, not because it makes us safer, but because it creates the impression of making us safer and that they are "doing their job." 

Then, there's that nagging "city-center-to-city-center" argument.  Yes, airports are, more or less, out of town.  It takes time to get there.  In "old Europe," the train stations are in the center of downtown along with most businesses.  Also, you don't need to rent a car in Europe after your train ride because there are superb urban and regional transit service connectors, as there are from their airports.

However, outside of Manhattan, you cannot say that the major cities of the US -- and certainly not the Bay Area and the LA Basin -- are well networked with urban and regional transit service.  So, what's my point?  The plane ride will still beat the train, DOOR TO DOOR. 
The HSR promoters want you to believe that there is a magical Goldilocks zone (of 100 to 400 miles) wherein HSR is faster than flying or driving.  Even if that may be true theoretically, it's simply not true in practice.

Or, to be more correct, it is true in Europe or Japan, but it will not be true in the US.  We work and live far too spread out to make inner city to inner city sufficient terminals for our trips.  Without Europe's superior transit systems, we will still have to rent a car to get where we're going.  That's true in the Bay Area and it's true in the LA Basin. 

With all due respect, Mr. Krugman, I beg to differ.

Then, there's one more little point that I couldn't put my finger on. It's in the diagram.  Note that the train starts at the intersection of the time and distance line.  But, the plane has been given a handicap of a more distant starting point. That's Krugman's erroneous assumption, which we have rejected.  No, it does not take longer to get to or from an airport in either the Bay Area or the LA Basin.  Indeed, there are many airports in the LA Basin and there will be only one high-speed train station, Union Station, to get to or leave from. And that makes this diagram not an objective description, but a biased assumption.
The New York Times Opinion Pages
The Conscience of a Liberal
Paul Krugman
March 2, 2011, 5:41 PM
Trains, Planes, and Automobiles
Some of the comments on my various pro-train posts have been along the lines of “Oh yeah, try taking the train to Los Angeles.” But that, of course, misses the point.

I think about the trains/planes comparison something like this: planes go much faster, and will continue to go faster even if we get high-speed rail; but there are some costs associated with a plane trip that can be avoided or minimized on a rail trip, and those costs are the same whether it’s a transcontinental flight or a hop halfway up or down the Northeast Corridor. You have to get to the airport at one end, and get from it at the other, which is a bigger issue, usually, than getting to and from train stations that are already in the city center. You have to wait on security lines. You have to spend more time boarding. So if we look just at travel time, it looks like this:

Suppose that I put those fixed costs at 2 hours; suppose that planes fly at 500 miles an hour; and suppose that we got TGV-type trains that went 200 miles an hour. Then the crossover point would be at 667 miles. It would still be much faster to take planes across the continent — but not between Boston and DC, or between SF and LA. Add in my personal preference for train travel, and I might be willing to train it to Chicago, maybe, but not to Texas.

Now, if we got vacuum maglevs

At short distances, there’s a somewhat similar crossover between trains and cars, having to do with whether they run from and to the desired destinations. I will and have taken the train to Newark; I drive to New Brunswick.

But the point is that there is a niche that could be served by fast trains.