Relative cost of putting in transit infrastructure before you build vs after?

I’ve been reading about some of the major urban transit initiatives and how mind bogglingly expensive they are and it seems like a huge bulk of the cost comes down to essentially performing surgery on a city while it’s still breathing. Major traffic disruption is politically untenable so workers are forced to work around the needs of residents.

It made me wonder, has any city considered speculatively pre-building transit infrastructure such that when the density increases to the point that it’s needed, it can be placed in far more cheaply?

Examples of this might be using cut and cover to bore out speculative tunnels which can either become highways or subways when the time comes. Our reserving large strips of land and turning them into parks or some other low value land such that they can be torn out and a highway or light rail system gets installed.

The main determination of how feasible something like this would be is down to how much cheaper it is to build the transit before the city rather than vice versa. If it’s hugely cheaper, then it doesn’t matter if even if the majority of the speculative transit infrastructure never gets used, it still ends up a better option than trying to graft better transit onto an existing city.

American history shows us several models of pre-building mass transit.

In the 19th century, commuter railroads were built out from most northeastern cities. Railroad suburbs are somewhat hard to spot today, as the auto allowed homeowners to fill in the surrounding spaces. Probably the noticeable example is the Main Line of Philadelphia, which spawned a string of expensive suburbs.

Lower income groups also sometimes benefited. The IRT subway was extended to The Bronx in 1908 and was conceived as a way of lessening the congestion in the Lower East Side by providing access to Manhattan from newer affordable housing in The Bronx. That was extremely successful. Population in The Bronx doubled from 1900 to 1910 and again in 1920.

Why did these work? Two reasons. People knew that populations were booming and they knew that they had only very few places to grow into given the technology of the time. Why haven’t they been copied more in the U.S.? Same two reasons. Places that need to retrofit mass transit are not booming in the same way and they don’t have empty spaces that will reasonably draw new housing.

The booming cities are in the west and southwest, areas that were shaped by the automobile. The auto is the perfect carrier for low-density areas. Mass transit, conversely, requires high-density nodes of destinations somewhere along the line. Two million people come into Manhattan every day. Phoenix or Las Vegas or Albuquerque has no such downtown. Even if the land is cheap, ridership on any given route is low and hard to make economical. Phoenix does has a 20-mile rail line connecting several nodes. It’s successful, but was built in 2008 after the nodes existed. It couldn’t have come first.

Places that are booming and expanding in today’s world exist, China being the prime example. It’s problem is that it can never get ahead of the population. A million people can move to a city before a subway can get finished. It’s always playing catch-up.

Here’s one more thing builders have to consider. Train lines and subways are essentially forever. Nobody will ever move the Philadelphia Main Line just because population drops or because a need exists elsewhere. The New York subway system feeds The Bronx. Growth is taking place in Brooklyn and Queens. Availability of transit is not the only determiner of where population goes. Guessing the future is a losing proposition. It will suck your funds dry before enough of your bets come in.

There’s also the psychological factor. Nobody wants to spend their money, regardless of whether they’re taxpayers or investors. So it’s hard to convince people to spend money now for a need that you anticipate will exist in ten or twenty years.

I’m pretty sure it’s done in many new developments in China, particularly newly built satellite cities that expect to be linked to or absorbed in to the main metropolis sooner than later.

The I-105 freeway in Los Angeles was built with an extra-wide median specifically so that a rail transit line could be built down the middle later. That line was in fact later built, it is now the Green Line.

I’ll just point out that most surgery is performed while the patient is still breathing.


It would probably be cheaper if they didn’t have to.

Let me just point out that the main line of Philly was also the main line of the PRR from Boston to Chicago and St. Louis to which they added a dozen commuter stops. More interesting were the other commuter srail lines, to Chestnut Hill and Trenton West, for example. And the dedicated trolley rights of way to Sharon Hill, Media, and, once upon a time, to West Chester, the P&W line to Norristown, and so on. Were these built before there was a demand?

Then there is the commuter rail in Montreal. The one I use regularly was definitely built on a “if we build it, they will come” basis and led to a string of residential suburbs and so is a successful example of pre-building. But I think it started as a way for a third rail corridor in Canada (Canadian Northern Railway) that was abandoned. I think the same was true of the Norristown line outside Philly.

Not only that, but tunnels and other infrastructure need constant maintenance. I imagine it would be even harder to get people to pay for maintaining a “tunnel to no-where” than to convince them to pay for the initial construction.

Good post, Exapno Mapcase. My thought is transit systems are a response to regional demand, for the most part.

Some places have sort of done this inadvertently, by taking over older, disused rail lines and the like, either for light rail projects, or for toll-road/highway traffic. That way, they don’t have to rip too mcuh stuff out and/or buy a whole lot of property, etc…

Witness the DART rail system in Dallas- almost all old disused commercial railroad lines, and the Westpark Tollway in Houston- primarily run where an old disused commercial rail line once was.

Erastus Wells built a streetcar line that ran along the St. Louis city limits out to the (then) distant towns of Florissant and Ferguson. For good measure, he then built a town on the line and named it Wellston.

Of course, that was a private individual and a private development. Also, the streetcar companies in St. Louis had a bad habit of going bankrupt almost constantly.

When they built the first subway under Yonge St. in Toronto, at the station at Queen St. in 1948, they built a platform underneath for the future crosstown Queen St. subway in the heart of downtown. the platform still sits empty, walled up behind the walkway underpass between the two platform sides of the old subway. Politicians being politicians and the dithering procrastinators they are, while they have built a Bloor Subway (1.25 miles north) and an Eglington LRT (3.75mi north) and Sheppard (7 miles north) they still are discussing whether a downtown crosstown subway is needed and when. Meanwhile, all those other lines feed a Yonge subway so congested now that all that is missing is the train packers like they have on the Tokyo subway.

It’s not that the earlier engineers guessed wrong, it’s that later politicians were so cheap and vision-free.

At least in the US, the integration of land use with new rail lines has not worked very well in recent decades. The neighbors along BART or the Washington Metro, and even along Portland’s MAX line, have fiercely resisted allowing denser development around the train stations. The combination of first building transit lines where there’s cheap right-of-way (like old freight railroads) instead of where it would serve the most residences and then never getting around to densifying the areas surrounding the new stations means that for the last 40 years we’ve just been building really expensive peripheral parking lots with fancy shuttles to take rich people to their downtown offices.

I think there are current instances of reserving land for later infrastructure. This costs less while waiting for people to start complaining about transit after they have moved in. You can also more easily justify the cost, and have an in place tax base that will be using it. Just don’t let those areas become too nice in some way. Or someone will complain when it is converted to transit.