The Obama administration recently pledged $8 billion for high-speed rail. While just a fraction of the overall stimulus package and just a drop-in-the-bucket of what is needed to build a real national high-speed rail network, the funds generated considerable hub-bub and, for some, outright jubilation among regionalists, environmentalists, energy efficiency advocates, and those who have long fought for improved U.S. rail transit. It also has encouraged a mad political scramble for funds as regions position for federal monies. In Canada, there is a mounting drumbeat for high-speed rail connecting Windsor, Toronto, Ottawa, Montreal, and Quebec cities and also for connecting Vancouver to Seattle.
For starters, here’s a map of proposed U.S. high speed rail projects:
It’s clear that the U.S. and North America lag far behind countries like Japan with its Shinkansen or France with the TGV on high-speed rail connectivity.
But how to base decisions on what routes get funded? How to avoid a purely political outcome and create a framework for investing in high-speed rail that makes the most economic sense?
Here’s a map of North America’s mega-regions:
The largest of North America’s mega-regions is the great “Bos-Wash” mega-region initially identified by the geographer Jean Gottman. It stretches down the east coast corridor encompassing the east coast cities of Boston, New York, Philadelphia, Baltimore, and Washington, D.C., and is home to more than 50 million people and produces more than $2.2 trillion in economic activity. Its economic output is greater than that of the UK and France and more than double that of India or Canada. The second biggest, which Gottman dubbed “Chi-Pitts,” covers more than 100,000 square miles and is home to 46 million people, producing $1.6 trillion in economic output. Taken together, America’s mega-regions produce more than three-quarters of its economic output and the lion’s share of its innovations (see the table below):
In the main, the proposed routes map pretty well to U.S. mega-regions. Given the fact that megas are dense and interconnected centers of population and economic activity, it makes sense to develop high-speed rail connections within mega-regions first, and later develop connections between contiguous ones, say for example down the east and west coasts or across the Great Lakes region.
The table below, compiled by Patrick Adler at the Martin Prosperity Institute, shows the distance between key cities and then compares the driving times (calculated on Google) to current top high-speed rail speeds (from Transportation Quarterly):
Philadelphia becomes a veritable suburb of NY, its commute time shrinking from nearly two hours to slightly more than a half hour. Washington-NYC and Boston-NYC become hour-and-a-half trips. San Diego becomes a bedroom suburb of Los Angeles. And commute times shrink considerably across Cascadias’ main cities: The time to get from Portland to Seattle shrinks to just over an hour, while travel between Seattle and Vancouver is reduced to less than an hour. It would take just slightly longer than an hour and a half to get from Charlotte to Atlanta. And commutes between Dallas and Houston and Dallas and Austin shrink to an hour and a half or less.
Better high-speed rail connections promise considerable economic efficiency gains. And they also promise to relieve the psychological burdens of commuting by car. Research by behavioral economists like Nobel prize-winner Daniel Kahneman finds that long car commutes are among the things that most adversely affect our happiness.
But there is an even bigger and more fundamental reason to connect our mega-regions through high-speed rail. As I recently argued in The Atlantic, our current economic crisis promises to powerfully reshape America’s geography. There will be winners and losers, and a new economic geography will emerge in time.
Geographic expansion, as I noted there, is a fundamental axis of economic recovery and development. Recovery after the Long Depression of the 1870s was in part powered by the rise of the large-scale industrial city that grew up around raw materials, ports, and railroads, expanding outward along its early street-car lines. While many see the rise of Keynesian spending (particularly World War II spending) as key to U.S. recovery from the Great Depression of the 1930s, post-war recovery was propelled by the rise of another era of geographic expansion - the rise of the Sunbelt and the massive growth of auto-oriented suburbia. Demand for cars surged to move workers between home and work. And suburban houses all had to be filled with the refrigerators, washing machines, dryers, television sets, and consumer appliances rolling off America’s assembly lines. This post-war auto-oriented “fordist” development model worked to ensure that mass production and mass consumption could grow together fueling the expansion of America’s great golden era.
But fordism has come smack up against its limits. It’s cheaper to produce many industrial goods off-shore, and the geography of post-war suburbia stretched to its breaking point. It may well be impossible for sustained recovery to come from breathing life back into the banks, auto companies, and suburban-oriented development model. A new period of geographic expansion – or what geographers term a “new spatial fix” – may well be needed to spur a renewed era of economic growth and development.
The history of capitalist development is the history of the more expansive and intensive use of space. Post-war suburbs, the rise of larger metropolitan areas, the development of multi-nodal regions with edge cities as well as downtown cores are part and parcel of this process of geographic development. It’s a mistake to consider suburban sprawl a backward step (as some do), and to see only more compact urban style back-to-the-city development as a path to the future. The rise of the mega-region is the cornerstone of a new, more intensive and more expansive use of space.
New periods of geographic expansion require new systems of infrastructure. Ever since the days of the canals, the early railroad, and streetcar suburbs, we’ve seen how infrastructure and transportation systems work to spur new patterns economic and regional development. The streetcar expanded the boundaries of the late 19th and early 20th century city, while the railroad moved goods and people between them. The automobile enabled workers to move to the suburbs and undertake far greater commutes, expanding the geographic landscape still further.
Mega-regions, if they are to function as integrated economic units, require better, more effective, and faster ways move goods, people, and ideas. High-speed rail accomplishes that, and it also provides a framework for future in-fill development along its corridors. Just as development filled in along the early street-car lines and the post-war highways, high-speed rail will encourage denser, more compact, and concentrated development with growth filling in along its routes over time. Spain’s new high-speed rail link between Barcelona and Madrid not only massively reduced commuting times between these two great Spanish cities, according to a recent New York Times report, it has also helped revitalize several declining locations along the line.
It’s time to start thinking of our transit and infrastructure projects less in political terms and more as a set of strategic investments that are fundamental to the speed and scope of our economic recovery and to the emerging shape of the economy, society, and communities of the future.







May 5th, 2009 at 7:55 am
What about a train that goes from California to Texas to Florida, Northern California to Pheonix to Denver to New York, Washington to Denver to Chicago to Boston in 3 to 4 hours. We should have three high speed parallel lines along the northern, central and southern regions that span the entire country. If we’re going to step up to the plate in this effort, let’s do it right. These smaller lines are great; however what about the people who want to travel across the country? Quit wasting tax payers dollars unless you’re going to add value to the system. This is a great project and will add millions of jobs across the country. Why not do it all today and become a mass transit powerhouse. I would use this system just as frequently as air travel.
May 5th, 2009 at 8:14 am
Problem with your driving times table.
Toronto to Windsor in 8:28? Via Cleveland, sure. But who would drive that? Makes no sense. Windsor to Toronto via normal driving (the 401) is 4 hours, which provides not much of a time savings over high speed rail (which I love).
A more relevant high speed rail line is Chicago-Detroit/Windsor-Toronto-Montreal. Buffalo could be a stop on a New York-Toronto line.
May 5th, 2009 at 8:20 am
Living in a town where two of the proposed high speed rail lines are supposed to go through (Chicago-Detroit and Chicago-Indianapolis), this will never, ever happen.
This country’s days of big builds are over. It is simply too burdomsome to go through emminant domain to get the land needed for these projects.
Sure, you could just live with the current system where passenger trains share rails with freight, but that’s not high speed rail. Unless you get passenger-only lines, real high speed rail is not going to happen. To do that, you really need to just blow through certain neighborhoods, like they did with the interstate system, and that’s just not going to happen anymore.
Also, Chi-Pitt? That’s an unfortunate monkier.
May 5th, 2009 at 9:31 am
Buzzcut – “Chi-Pitts” is the name Gottman originally gave. I kinda like it.
Sean – Nice catch. I should have caught that myself. My wife and I drive it in less than four hours regularly. The US-centric google our team used calculated Windsor Vermont! It’s corrected now.
May 5th, 2009 at 2:14 pm
First of all, I have to say: great post. I really was looking forward to read something like this about high-speed rail in North-America and economic regions.
But just one point, about Spain. While is true that Spanish high-speed rail reduced commuting times between the big cities (BCN-Madrid, but also Madrid-Sevilla, and Madrid-Valencia is coming), and helped economic relations between them, is not so clear how is it affecting to medium-sized cities where it stops. As you know, a significant increase in accessibility to a location also means a significant increase in ease of leaving (for living, consuming, cultural life, whatever).
It mostly depends on the ability of medium cities to become more attractive, and its previous economic and social context. At this point, there’s no substantial proof of high-speed rail “helped revitalize several declining locations along the line” Madrid-BCN, maybe because is too early, although City Hall Administrations of this medium and small cities are trying to make the most of the situation. For example, medium and small locations on Madrid-Sevilla line, the first high-speed train in Spain (1992), haven’t been able to take advantage of it: in the best case, they didn’t get worse than before. In the worst case, Madrid had too much attraction power.
May 5th, 2009 at 6:29 pm
High-speed rail is a great connector and productivity producer. Starting with the feds map, where does it make sense to build rail?
The competition is airplanes and cars. So where does rail make more sense than flying or driving? When its easier, cheaper and more convenient. Generally major cities that are more than 100 but less than 500 miles apart.
For example, currently its about as fast to drive Portland to Seattle as to fly, 3 hours either way — by the time you drive to the airport and park, get there an hour early, wait in line, fly 1/2 hour, collect luggage, then get from the airport to downtown. So most people drive, a generally unpleasant trip when you can’t do anything else. Decent train service, never mind high speed, would be competitive.
So my opinion is good service on the main routes. • Bos-Wash obviously, its a chain of centers that already has OK rail. • Cascadia (of course I’m prejudiced) it would tie the region together, including B.C. • Minneapolis-Chicago-Pittsburgh-NY. • SF-LA-San Diego is a little far and flying is an option, but it’s 10% of the nation’s population. (I would build the inland route, avoid the Coast Range and fault lines- whether it would improve the Central Valley, I don’t know.) • Probably Chi-Detroit-Montreal-Toronto, connect to Boston/Toronto.
The others I don’t know. I don’t know if Southern Ohio/Indiana/Kentucky have enough going on, and someone who knows the South better than I would have to say about them.
The stimulus isn’t going to create Japanese-French style bullet trains. But developing good rail service will lay the groundwork for them next decade.
Buzzcut, you’ve identified a problem and its solution. Built the rail routes along the Interstate right-of-ways. You’d have to widen them a bit, but the adjacent properties are already devalued.
May 6th, 2009 at 12:35 pm
Interesting, but you need to connect Houston-Dallas-Austin more than you need to connect Houston-New Orleans. Also, I think it would make sense to connect Denver with OKC and Texas because of energy – particularly since it is not connected anywhere.
May 7th, 2009 at 10:14 am
Given the relative volumes of passenger and freight on rail, would it not make more economic sense to develop high speed freight rail?
May 10th, 2009 at 10:26 pm
There are several other mistakes in regards to the high speed rail timetable which Sean did not mention. First, and the most obvious, is the title. High speed rail timetables? I only notice one so why use the plural. Second, the chart could really use a legend. Units of measurement for time and distance would be beneficial; there is no such indication. Third, when I used Google maps to verify some of the times and distances listed in the chart, I found slight inaccuracies with them all. Also, in regards to the Toronto-Windsor route information, the distance listed is 229.90 miles (which is correct according to Google); however when you do the calculation based on the TVG speed of 155 mph with the TVG travel time between Toronto-Windsor (1 hour and 47 minutes as listed in the chart) then the distance between the two cities would be 276.4 miles. Where is this discrepancy coming from?
Overlooking all of the factual inaccuracies, the aesthetics of the table are a nightmare. It is fragmented, the headings are redundant, and the outline is a mess.
I would have to say that it is pretty unforgivable to have so many mistakes in a single chart. It certainly takes away from the credibility of the article when the informative chart displayed is riddled with inaccuracies.
January 30th, 2010 at 10:53 am
Richard, can you repeat this analysis with the additional routes that have been approved?
I’m particularly interested in the impact of the Cheyenne/El-Paso HSR route on the Denver and Albuquerque regions.
January 30th, 2010 at 10:54 am
Sorry, I forgot to link to the new routes: http://www.fra.dot.gov/us/content/2243