Where Would Hyperloops Work?

The viability of Elon's Hyperloop may depend on achieving a "goldilocks zone" for distance.
By Ethan Finlan | Jul 09, 2018 |
By Ethan Finlan | Jul 09, 2018 |
The Hyperloop at Launch Festival 2016. / Flickr

The Hyperloop is the theoretical transport innovation that urbanists love to hate.

Conceived by Elon Musk in 2013, the system in theory would transport passengers at speeds as high as 700 or 800mph, through high velocity tubes. While this would be revolutionary, much of the coverage of the Hyperloop from urbanist circles has been critical. Speeds achieved in testing thus far have been lower than the 700mph limit. More fundamentally, hyperloops would have lower per-vehicle capacity than most trains or even buses. There are concerns over safety and rider comfort. Politicians have grown enamored with the prospect of hyperloops, since it would let them waste money on a speculative technology while avoiding more tangible improvements. For his part, Musk has taken a somewhat sweeping approach to the project’s benefits. He's dismissed California’s high-speed rail (which, granted, has its problems, and they are serious) as fundamentally outdated, while touting an untested technology, and brashly dismissing informed criticism of his proposals.

For the record, I share most of these concerns. As Railway Gazette notes, to approximate the capacity of a two-track high speed rail line, 23 50-seat pods would have to operate every 9 seconds hourly, likely requiring increased infrastructure. Alon Levy, an early critic of the technology, observed that Hyperloop One estimated the cost savings to be a less dramatic 2/3 over high speed rail.

The most frustrating aspect of this discussion has been the the exaggerated claims made by both sides: On one hand, boosters overplay their hand by definitively presenting Hyperloops as rendering high-speed rail obsolete; on the other, any discussion of the technology seems to result in scorn from some urbanist circles and, occasionally, an assumption of bad faith.

With the aim of solving problems, then, let’s ask: Where would Hyperloops work, and of the candidates, are there routes that would be better, or worse, than others?

In Levy's article, two fundamental questions are posed regarding Hyperloop technology:

The economic hitch is, what is Hyperloop for? The technology suffers from tension between two opposing forces. The first force is speed: as a very fast technology, Hyperloop is the most useful for long-distance travel. At the distance of Musk’s original Los Angeles-San Francisco idea, security theater and design compromises about station locations…would eat up the entire travel time advantage over conventional HSR. At longer distance, such as New York-Chicago, Hyperloop would still win on time … The second force is that Hyperloop still requires linear infrastructure, so it becomes less cost-effective versus planes as the distance increases.

This is a valid contention, which we can use to develop a theory of where super-speed ground transportation would work. Put simply: if the distance is too short, Hyperloop isn’t viable because of the poor ability to serve intermediate destinations without additional infrastructure – road, rail, and bus have the advantage here. If the distance is too far, Hyperloop isn’t viable because of the higher infrastructure costs that air travel doesn’t have to deal with. For Hyperloop to be viable, there has to be a “Goldilocks zone” of not too close and not too far. Though this is limited to American examples for the sake of time, I believe that several other lines under consideration – such as the one being built between Abu Dhabi and Dubai – would add backing for my theory.

Another condition: there can’t be demand for too many or too close intermediate stops.

On that basis, it seems valid to conclude that the proposed New York to DC line is more dubious. A nonstop service would get use, to be sure, but the comparatively short interstation distance raises questions. Rail, by contrast, has a lower stopping penalty, and can support more service patterns with, generally speaking, less infrastructure.

New York-Chicago, then, is more plausible. But it’s still on the far side, and with flights throughout the day between each city, it’s hard to wonder if flying isn’t more scalable.

But what about meeting in the middle?

In addition to Chicago, there are numerous mid-sized cities where there is a good amount of demand for service from the Northeast. But because the demand is not quite as intense, flying is more expensive. A fast link to the Northeast would arguably provide agglomeration benefits. Fast service to a point from which several of these cities could be reached would provide a critical mass of demand, one which could be competitive, and require lower infrastructure costs.

Consider that a Hyperloop could be built from New York to Pittsburgh, with feeder rail and bus services on the western end linking it to Detroit, Cleveland, Cincinnati, Louisville, Indianapolis, and Charleston. If the trip to this “Midwest Hub” could be made in 30-45 minutes, even with the transfer to the slower mode, the trip becomes competitive with flying. Intermediate stops could be made in Harrisburg and Allentown. The line could be built along the Pennsylvania Turnpike. Terrain, of course, would be a challenge here.

The cities of the southern Midwest and Southwest regions, respectively, would be a good fit for super-fast technology. Here the issue is not so much infrastructure costs as it is demand – how much demand is there to go from, for instance, Denver to Kansas City? Enough to justify the high capital cost of Hyperloop? On the other hand, the Texas Triangle, and routes from Los Angeles to Phoenix and Las Vegas, may be more viable.

Theoretically, then, there are markets where Hyperloop makes sense, although they are somewhat narrow - routes where either there are a) two nodes with a critical mass of demand for service between them, which can bypass intermediate points, or b) for connecting service, enabled by a faster trip. Much remains to be answered about this prospective technology – such as whether it can carry enough passengers per vehicle to make it viable – but an all-or-nothing approach only crowds out an understanding of the possibilities.