The American Autobahn or: How I Learned to Stop Worrying About High-Speed Rail
The window for high-speed rail is closing fast. It’s time for a closer look at what’s going to emerge in its place.
I’ve been a train geek for a long time. I play train and transit games obsessively. I love model train layouts and visit as many of them as I can. From a distance I’ve admired the bullet trains of Japan and and the maglev trains of China. I’ve taken the Eurostar from London to Paris and the sleeper train from Amsterdam to Munich. I know firsthand that rail travel can be amazing.
But I’m also a realist, and it seems to me that it’s time to face facts about high-speed rail in America. Here’s a fact: passenger rail travel isn’t awesome in the US, and it’s not going to be awesome anytime soon. Maybe not ever.
As of right now, there’s exactly one HSR line under construction: the line from Los Angeles to San Francisco. It should be operational in 2029 if everything goes really, super duper well. California is unique among American states in its political ability to finance such a project (“because Jerry Brown said so, that’s why”); elsewhere, HSR projects continue to languish on drawing boards and in the twin purgatories of political tugs-of-war and environmental impact studies.
That’s because, leaving aside California, the US just isn’t like those other countries where rail travel is awesome. We’re not France or Germany or Japan, where closely packed cities make train travel ideal for the majority of citizens. We’re not China, whose government can simply decide to spend huge amounts on infrastructure to connect widely dispersed cities, without the inconvenient necessity of placating voters and taxpayers.
No, the US is a sprawling country of individualists, and our people are spread all through our great big beautiful land. We’re a nation primarily of suburbs, which means that mass transit only works well in our largest and densest cities — i.e. where the vast majority of us don’t live. Our decentralized form of government gives our rural citizens as much of a say as our urban citizens in how we raise and spend federal money. And while some corridors of urban citizens continue to be very supportive of high speed rail, our rural population has absolutely no reason to support rail at all, high speed or otherwise, because rails don’t come anywhere near them.
But roads do.
American geography, American politics, and American temperament have conspired to make America the land of the automobile and the road. We are, and always have been since the first Model Ts rolled off the line, a culture of highway-goers.
Which is why the next logical step for American transit is not high-speed rail. No, our future is the 21st century American Autobahn.
Imaging the American Autobahn
I know. “American Autobahn, what the hell nonsense is that?” Fair enough. It’s just a fancy way to talk about the coming grid of autonomous vehicles (AVs). I like the sound of it, what can I say?
Let’s imagine this American Autobahn.
Imagine it’s 2030, and you need to go on a business trip. Let’s say you live in… *spins wheel of random American cities*… Ames, Iowa. And you need to get to… *spins wheel of random American cities*… Austin, Texas.
Now let’s imagine you’ve got a 9am meeting next Tuesday morning with a client in Austin. Oh, and for extra fun, let’s say that you found out about this meeting on Friday afternoon. Thanks, boss!
How would that trip have worked in 2016? Let’s travel back to those grim days for a moment.
Because all sensible people hate air travel during the Reign of the TSA, you first look at your other travel options.
Ha ha, no, just kidding, you don’t do that! You don’t even consider doing that for a second, because there are no other reasonable travel options. Amtrak? Please; there’s not an Amtrak station for a hundred miles. You could ride a bus for 22 hours, but that’s a dingy and dismal experience, so that’s out. You might consider driving if you really, really, *really* had to — but that’s a 14 hour drive each way, which would mean saying goodbye to Monday and Wednesday.
Air travel it is, then. The closest real airport is Des Moines, and according to Google Maps, it’s a 48 minute drive without traffic. Orbitz has flights from DSM-AUS for about $600 — not a bad price on three days notice, actually. Nothing direct, though, and nothing leaving early enough on Tuesday morning, which means you look for a Monday night flight, and you find one that leaves DSM at 7:11pm, connects thru DFW, and arrives at AUS at 11:14pm. Your meeting will be about two hours, which should leave just enough time for lunch and then back to AUS for the 2:20pm return flight to DSM, arriving at 6:41pm. That’ll put you home by 8pm or so, just in time to watch The Sporting Event of Interest.
You book the flight. Ordinarily you’d book an earlier flight, because as an experienced business traveller, you know to avoid later flights because flight delays tend to stack up over the course of a day — but you’ve got other meetings on Monday afternoon that you can’t get away from. So you’re stuck with the late flight out. You cross your fingers, and you book it.
You also book a hotel in Austin for Monday night, and you plan to take a cab from the airport on Monday night, and then an Uber back to the airport on Tuesday afternoon.
You also know that you’re going to have a presentation to finish, but you figure you can do that from the airport/plane/taxi. A few good hours of working time? Maybe, assuming no crying babies and enough elbow room to work. Again, fingers crossed.
Total travel time, door-to-door, from leaving your front door at Ames at 5pm for the hour-ish drive, and stepping out of the taxi at your hotel at 11:45pm: six hours, forty-five minutes. With a roughly equivalent travel time home when you’re done.
Total travel expense: $600 for the flight, $150 for the hotel, $50 for the taxi/ubers to and from the airport. $800 total.
And that’s the best case scenario in 2016. Needless to say, there’s lots and lots of room for that best case scenario to go completely sideways — in fact, if that Austin meeting is *really* important, you’ll seriously consider leaving Ames early on morning anyway to make sure to Austin on time, and to blazes with the Monday afternoon meeting.
Okay! Back to 2030.
You go to Orbitz. Next to their “Flights” tab, there’s the “Cars” tab. Just like in 2016 — except that in 2030 “Cars” means, almost exclusively, autonomous vehicles. Everyone still hates airports in 2030, and you’ve heard good things about some of these long distance car rides, so you click the “Cars” tab first.
You put in your home address (some subdivision outside of Ames), your destination (an office building in downtown Austin), and when you need to be there (9am).
Orbitz recommends the Hertz/Lyft Easy Sleeper: a single-seat vehicle that’s roughly the size and shape of a first-class airplane seat, encased in a super-safe vehicle shell. Electric, of course. The chair is super-comfy, has an excellent harness for your additional safety, and reclines fully for sleep. You’ll also have 7G gigabit wifi and a 42" monitor. Direct travel time from Ames to Austin: 12 hours, including two scheduled refueling stops of 10 minutes each. Average travel speed, including stops: 80 mph. Fast, right? Of course! It’s made possible by the new dedicated high-speed AV lanes on I-35 — which is, let’s be honest, the speed that most human drivers were traveling on I-35 anyway. You’ll slow down to 60-ish around KC and OKC and Fort Worth, but you’ll speed along at up to 90–100 mph along the rest of the route. You decide to take an extra hour, giving you the freedom to stop along the way for a bite to eat and a bathroom break. You select a 7:45pm pickup time on Monday night, which will put you in Austin in plenty of time for the Tuesday morning meeting at 9am.
Your meeting in Austin is scheduled to wrap up at 11am on Tuesday, so you schedule return pickup at 11:15am, which would put you home a little after midnight. You don’t have to worry about missing The Sporting Event of Interest; that’s what the 42" monitor is for. Maybe you’ll even bring some beers along. It’s not like you’re driving, ha ha ha!
Good enough. You book, and you’re done. No plane, no hotel room. Just an awesome autonomous travel pod. You don’t even bother to check the airline option, since it won’t be nearly as comfortable or as flexible — and if the meeting goes well, you may need to extend your stay, and rebooking the flight would be an incredible pain in the ass.
Cost: $0.50 a mile, standard for the high-end Easy Sleeper. You could get the El Cheapo with a fixed vinyl seat and no TV for $0.35 a mile, but why? It’s a big meeting, and the company is paying; treat yo’self. Cost: $1000 total (or $700 if your boss is a penny-pinching bastard and forces to you go cheap).
Along the way — and this is the amazing part — you do exactly what you want to do. You work on your presentation, you sleep, you watch a movie, or you play video games. Maybe you even read a book. A real one, with pages and stuff! So retro! You have 13 uninterrupted hours to spend any way you want. Or, if you want to stop somewhere along the way to stretch your legs, that’s ok too — you just pay a little more to have your AV for a bit longer than you’d originally planned. If you need a bathroom break, you stop and take one. If you need some fresh air, you stop and get some. Say, don’t you have a friend in Dallas you haven’t seen for a while? Re-route and stop by for dinner on Tuesday night. You can watch the Sporting Event of Interest Together! And then get back in the AV, and reprogram it to take you directly to your office in Ames in the morning, and grab some zzzs.
Yes, the air travel option is still technically faster. But the total time of the trip — Monday night departure, Tuesday night return — is comparable. And besides, who the hell cares? Air travel sucks. Anyone who does it regularly hates it. We have come to despise air travel with a passion, and with good reason. It’s uncomfortable, it’s dehumanizing, it’s inconvenient, it’s brutal on the environment, and it wins only because there’s no option that’s even remotely comparable.
The American Autobahn will become that option. What’s more, it will happen with shocking rapidity.
Comparing Air Travel, High Speed Rail, and the American Autobahn
Now let’s imagine that A Miracle Happens, and America suddenly decides that High Speed Rail is worth the trillion dollar investment. It’s 2030 again, and we’ve brought the USHSR map to life.
Lucky American consumer! You now have three excellent travel options to choose from: Air, HSR, and AV. Let’s compare the transit times for these three modes of travel.
Let’s also make some simplifying assumptions in our comparisons. These assumptions may be imperfect, but I think they’re in the ballpark — close enough to give us an idea.
Assumption One: let’s assume that all of our trains will average 180 mph, which is comparable to the Eurostar currently. We’ll also assume optimal direct-line routes to simplify the math — even though for longer routes, this is almost certain not to be true.
Assumption Two: AV trips will be door-to-door, but HSR/Air trips are door-to-hub-to-hub-to-door. To account for this Last Mile Travel (LMT) problem, we’ll add half an hour to all HSR/Air trips — although depending on the distances from hubs to endpoints, the time could be shorter, or longer (as in our Ames to Austin example.)
Assumption Three: Airlines have Airport Time (AT) overhead for security screening and boarding, so we’ll add an hour to all airline flights.
Assumption Four: As AV technology continues to improve, the maximum speeds possible will go up. We’ll provide two AV times for reference: an 80 mph average (which people are doing with Teslas in semi-autonomous mode, right now, today,) and a 100 mph average (a guess at an upper-end capability of current infrastructure, above which we can assume that significant AV-specific road investments like banked curves, etc. will need to happen.)
Boston to NYC, 216mi
- Drive: 4h22m.
- HSR: 1h15m + 30m LMT: 1h45m.
- Air: 1h + 1h AT + 30m LMT: 2h30m.
- AV80: 2h45m.
- AV100: 2h15m.
Dallas to San Antonio, 270mi
- Drive: 4h.
- Air: 1h + 30m LMT + 1h AT: 2h30m.
- HSR: 1h30m + 30m LMT: 2h.
- AV80: 3h.
- AV100: 2h40m.
LA to SF, 388mi
- Drive: 5h30m.
- Air: 1h15m + 30m LMT + 1h AT: 2h45m.
- HSR: 2h20m + 30m LMT: 2h50m.
- A80: 4h45m.
- A100: 3h55m.
Washington to Atlanta, 630mi
- Drive: 9h30.
- Air: 1h40 + 30m LMT + 1h AT: 3h10.
- HSR: 3h40 + 30m LMT: 4h10.
- AV80: 8h.
- AV100: 6h20.
NYC to Chicago, 789mi
- Drive: 12h30.
- Air: 2h30 + 30m LMT + 1h AT = 4h.
- HSR: 4h20 + 30m LMT = 4h50.
- A80: 9h50.
- A100: 7h55.
NYC to SF, 2904mi
- Drive: 42h.
- Air: 6h + 30m LMT + 1h AT = 7h30m.
- HSR: 16h15m + 30m LMT = 16h45m.
- A80: 36h.
- A100: 29h.
OK, let’s analyze the results!
- AVs will always win if other factors are equal. People like comfort, flexibility, and autonomy. AVs will always provide all of these in spades over planes or trains. For that reason, all things being roughly equal, we can assume that most people will choose AVs first, rail second, and airplanes last.
- For short trips of 300 miles or less, AVs will win. AVs are within half an hour of the best HSR/Air times, because Last Mile Travel time eats up any HSR/Air time advantage — which is why people frequently prefer to drive themselves in these scenarios in 2016. AVs will be the clear winner.
- For middle distance trips of 300–1500 miles, HSR might win, but AVs would also be a reasonable alternative to HSR or Air. HSR and Air speeds will be comparable, whereas AVs will be 2–6 hours slower. In these cases, the availability, comfort, convenience, and flexibility of AVs will still frequently win — maybe even usually — but when time is of the essence and the timetables and connections work well, people would choose HSR over Air.
- For long distance trips of 1500 miles or more, air will usually win, but AVs will still come out ahead of HSR. Sure, AV is even longer than HSR in these scenarios, but AV has a clear advantage because every AV will also be its own private sleeper car. HSR will never be able to match that advantage. If it’s a matter of speed, people will always choose Air, and if it’s a matter of comfort, people will always choose a sleeper AV. HSR won’t be nearly fast enough to compete with Air, and it won’t be nearly comfortable or convenient enough to compete with AV.
The Feasibility of HSR Versus AV
We’ve made a key assumption in the above scenarios: that Americans will actually commit to building a High Speed Rail network.
That is an extraordinarily generous assumption.
More bluntly: it’s a fairy tale.
Comprehensive high-speed rail can only exist in America if we commit to a gargantuan investment of taxpayer money — and such an investment, even if started now and aggressively pursued, wouldn’t provide any service at all, to anyone, for twenty years.
Beyond that up-front cost, taxpayers would have to subsidize the operation of this HSR network for another several decades — remember that Eurostar didn’t make its first profit until seventeen years after its launch, and even the Shinkansen, the model for the entire world, struggles to keep most of its lines profitable.
Getting our current Federal Government to commit to that kind of speculative investment would take a miracle — and as we all know too well, American lawmakers aren’t in the miracle business. They’re barely in the lawmaking business.
In the meantime, they can simply wait and watch as the American Autobahn takes shape. Every major car manufacturer now expects to have AVs ready for customers by 2020. That door-to-door sleeper ride from Ames to Austin is, essentially, five years away. It may not be as cheap or as comfortable as in our little scenario — but it will be *reality*. You’ll be able to get in your car in Ames, push a few buttons, go to sleep, and be in Austin twelve hours later. (OK, you might have to wake up once or twice to recharge your car if fueling stations aren’t automated yet.) And it won’t take a gigantic government subsidy to make it happen; it’ll be driven as a result of good ol’ American innovation and competition.
The competition in the AV space will be fierce, and breathtakingly innovative — and it will, in turn, make it easy for lawmakers to justify voting for highway improvements for AVs. Politicians already understand how to raise and spend highway money in this country; increased standards for AVs will just mean raising and spending a bit more of that money, except with much clearer and more immediate payoffs.
Not that it will be easy, of course. But it doesn’t have to be easy. It just has to be feasible — and a future dominated by AV travel is now not only feasible, but inevitable.
It’s time to stop fantasizing about high-speed rail in America.
(Except for California, maybe — good luck with the Jerry Brown Express. You’ll need it, because you’ll be competing with thousands of AVs moving up and down I-5 every single day.)