web analytics

Category / Transit

A blueprint for HSR on the East Coast May 26, 2010 at 3:53 am

Last week I stumbled across something I’ve been seeking for a few years now: a list of the 100 busiest air travel routes in the United States, complete with passenger numbers from March 2009. Not having anything else to do on that particular Tuesday, I set to work putting the numbers to some use and built several nice big charts, which you’ll find embedded at the bottom of this post.

If you find the spreadsheets overwhelming or just want the analysis, here’s what I found interesting:

  1. 24 of the 100 top air routes in the USA are between city pairs on the East Coast (specifically, the cities of Boston, New York, Philadelphia, Baltimore, Washington, Charlotte, Atlanta, Jacksonville, Orlando, Tampa and Miami; see maps at right).
  2. NYC-Miami leads the pack, with nearly 9 million passengers per year between the 3 NYC airports and airports in the Miami-Fort Lauderdale area. In the last 5 years this route has gained nearly 10% in passenger numbers. And 30% in the last 10 years.
  3. In total, over 60 million passengers travel by air between these 11 East Coast cities every year.
  4. Nine of these routes are between cities less than 500 miles apart (less than 3 hours travel time with the average speeds achieved by high speed trains in Europe and Asia); with a total of nearly 17 million passengers between them.
  5. Another 8 routes are between 500-1000 miles (including the fairly popular Miami-Atlanta [5 million/yr, 3rd nationally] and Atlanta to NYC [4.5 mil/yr, 5th nationally] corridors), with another 21 million passengers between them.

You can probably tell where I’m going with all this: building a less than 2000 mile High-Speed Rail system up the East Coast from Miami to Boston (with a small spur from Orlando to Tampa) has 60 million reasons why it makes sense. The skies in this part of the world are some of (if not the most) congested in the world, and instead of simply praising the Boston-Washington Acela Express “high-speed” train (which averages less than half the speed of what HSR trains in the rest of the world average) for taking 3 million passengers out of the skies and instead of working on a better air traffic control system (both of which are good things to do in general) we should be moving forward with another, much bolder, two-part plan:

  1. Upgrading the Acela system to State-of-the-Art speeds that rival those seen recently in Europe and Asia (imagine cutting the 4.5 hour trip from Boston to Washington down to only 2.5 hours (470 miles divided by 190mph average speed), or even shorter? Or NYC-Washington’s over 2.5 hour trip to under 90 minutes?) Could we bring the number of people who fly within this 500 mile corridor from 8 million people down to zero, freeing up terminal space at all the major airports in these 5 cities for more profitable long-haul flights? It was several years ago and I’ve lost the source, but I think I heard from a congressional hearing that this type of thing would cost ~$10 billion (or just over $20 million/mile for a doubling in speed). Not bad considering that the California HSR system is promising to cost at least 5 times that for a system less than double the size.
  2. Seriously pursuing the viability of building HSR from Washington D.C. to Atlanta and on to Miami. If a serious HSR system (a system on par with those currently being built and in service in Europe and Asia with average speeds of 180-190mph and cruising speeds of 200+ mph) would take even 20% of the passengers from the skies, that’s more people than currently ride the Acela Express every year and my guess, based on nothing but the Acela’s profitably success in the USA and the runaway success of HSR in the rest of the world (the Eurostar, for example, has basically killed the air traffic corridor between London and Paris while averaging a relatively modest ~135mph along the 300 mile route), is that, done right, an HSR system up and down the east coast could grab more than 20% market share (and this 60 million passengers per year market is nothing to sneeze at).

Of course, it’s that little phrase “done right” that’s the rub. Here’s my idea:

  • Super fast. There’s a difference between the definition of “high-speed rail” in the USA and the definition in the rest of the countries that have HSR. That difference needs to be addressed, and I think we need to adapt a policy and definition more in line with what Europe and Japan have (I would suggest raising our definition of HSR from 110mph to at least 150mph, which would unfortunately mean the Acela Express would no longer qualify, but that’s sort of my whole point). Aside from what we define it as, I think we need to be aiming to rival the speeds seen in Europe and Asia; meaning cruising above 200mph and average speeds of 170mph or more. The California HSR system is set to cruise at 220mph.
  • Involve the private sector. Yup. There’s a common argument against HSR that “if it were worth doing, some corporation would have already done it”. Ignoring politics for a minute and setting aside the whole big government vs. big corporations debate I have a straightforward answer to this argument: it’s folly to expect companies to put up the costs for rail infrastructure when every other infrastructure system in this country is government owned (I blogged extensively on this late last year), and with the exception of some of the farthest out libertarians, I think we can all see the sense in that. So here’s the idea: destroy the clause in Amtrak’s charter that says it has to approve every passenger rail system in the United States and bring private companies into the conversation about building rail systems (because frankly, I don’t believe that Amtrak knows any more than a lot of private companies how to do HSR right). Put up government dollars for the tracks and signaling systems and open up the role of operator to private companies (for Pete’s sake, Europe is even doing this). I think an argument against this is that it’s not guaranteed that any private companies would step up to operate trains. But I can see two ways around this: A) get a firm signed contract from a company prior to the start of construction, and/or B) have plans for a government-run system to step in if no private companies want to tackle the market. Also, nearly 20 years ago a group of companies agreed to do the entire thing (billion-dollar tracks and all) in Texas (because Texas only wanted HSR if they could avoid spending any tax dollars on it). The system never came about due to missed deadlines with the financing and opposition from Southwest Airlines, but that shows that it’s likely that basically offered “free” tracks (the biggest cost in getting an HSR system up and running) upon which to run a train business, a lot of companies would probably jump right in. Another possible private sector involvement is light cargo transport. If freight companies were included in the design phase and had tracks running right to their warehouses, who knows how many would opt to ship things via high-speed train rather than air, at least regionally. All of these uses could be charged (basically running the trackage system like a tollroad, only fair since it’s a premium transit option) and if managed correctly the government could actually come out ahead (or at least not totally behind) on their infrastructure investment.
  • Work with airports, not against them. In other words, downtown stations are invaluable, but stations at airports are too. This is something few HSR systems do, but which a lot of people have begun talking about and I think it makes a ton of sense. Not only are airports already linked into the local transit system (including car rentals, bus lines, taxis, etc.) this would automatically turn an HSR system into a feeder system for the airport system, possibly cutting out the need for shorter hop regional flights (the least efficient type of flight and one that loses the airlines money), and given the number of annual flights in America, I imagine that the “airport feeder” system would be quite a big market.
  • Code sharing. Related to the last two points I think it would make a lot of sense to work closely with airlines when setting up an operator(s) on the system and either invite the airlines to be rail operators, or at least encourage code sharing between the new rail operators and the existing airlines, this type of project is gonna be seen as a threat by the airlines, and rightly so; the best way to avoid a repeat of the Texas TGV-Southwest Airlines debacle of the early ’90s is to involve the airlines this time around, rather than working against them. Code sharing and/or trains operated directly by airline companies would make it possible to book a trip that takes you via train and plane all on one ticket/purchase, including luggage check-through, etc.  This is basically a win-win for everyone, except maybe anti-big business people who don’t care about having a seamless transit system (yes, that was sarcasm).
  • Express trains. I don’t have exact numbers as for how much time stopping at a station adds to an HSR journey, but with slowing down, speeding up and time spent sitting at the station (I would think at least 5-10 minutes for this) I imagine it adds up to half an hour or more. For this reason I would suggest every station on the route be “off-line,” meaning if an operator decides to run express trains and skip a certain station, but a different operator decides to stop there, they can both do that and the express train can just pass the “local” one by. The idea of express and local trains is obviously not a new one, but I think that having the off-line stations (perhaps even with several miles of off-line track for slowing down and speeding up trains) is pretty important. This would mean, for example, that if an operator wanted to try to capitalize on the nearly 10 million people who travel from NYC to Miami every year, but didn’t want to slow down the trip at all by stopping at any other stations, they could run a nonstop train and, depending on the train and track technology, could possibly do the 1600 mile trip at an average speed over 200 miles an hour, making it a nice, calmer, and not-too-much longer trip than by air (when you take into account airport security time, delays, etc.). This comes back to the idea of keeping competition open, as well: if all the stations are basically optional, that makes it more possible for a potential operator to mix and match what they want to do on the system.

Running the numbers for a quick (really, really rough) cost estimate based on how much the California HSR system is projected to cost per mile, and going off that number I’m remembering about how much it would cost to upgrade the Acela system to European speeds, we come up with upgrading the Washington to Boston trackage costing $10 billion (or ~$20 million/mile), and the other 1500 miles (Washington to Miami and Orlando to Tampa) costing roughly $56 million/mile (makes sense given that in Acela’s case land acquisition and most railbed work is already done) or a total of ~$85 billion, so around $100 billion for both the upgrades and the new. Now before you freak right out, keep a few things in mind: we’ve spent seven times that amount of money on the war in Iraq, and nearly three times that on the war in Afghanistan. I hate to play the “defense budget” card, but it’s really quite handy. The $100 billion for East Coast HSR would be spent over a period of perhaps a decade or more as impact studies are done, construction contracts are awarded, and the like. So let’s say we’re looking at $10 billion/yr for 10 years. The Department of Defense’s budget this year is $663.8 billion dollars. That’s one year. That’s 66.38 TIMES what we’d be asking for HSR per year. Either you think that every drop of that defense money is well spent, so then why would you care about an amount that’s only 1.5% of it? Or you think the defense budget is just too darned high, in which case I’d argue that we should skim some off the top and give it to HSR. Either way, I’m gonna argue for giving HSR it’s day, and starting with a transit corridor, that, it turns out, is quite a big market.

Edit (4am May 26): I just discovered that Florida received $1.25 billion for their recently revived HSR system from the Federal Economic Stimulus, this is said to cover half the cost of the Tampa-Orlando segment ($2.4 billion) meaning they’re budgeting it at ~28 million/mile, half of what California is budgeting for their system. At first blush I’d say the Florida number might be a little low because it’s several years dated, but I think it’s also possible that the California system could cost more per mile because of the more mountainous terrain in that state. All this to say that I now have another data point to base my EXTREMELY rough total cost estimate on, so it might very well be lower than the $100 billion number I came to above. Perhaps more along the lines of $70-90 billion. A quick look at a terrain map doesn’t show many or any mountain ranges that this system would need to cross, so my guess is that it could indeed be a fair bit cheaper than the CA HSR system. On the other hand, the Tampa-Orlando line is being built in the Interstate-4 median, meaning no land costs.