How to Get High-Speed Rail Back on Track

BY WILLIAM BOSTON

Forget flashy bullet trains and gazillion-dollar tunnels: Alon Levy, a fellow at the NYU Marron Institute of Urban Management, says America’s railways could reduce travel times and transform intercity trips with simpler, far cheaper fixes.

In a recent study, Levy and co-authors make the case that with a combination of smarter scheduling, new track switches that allow trains to approach stations at faster speeds and a fleet of modern electric trains, the Northeast Corridor connecting Boston to Washington, D.C., could be humming with near high-speed rail service. Imagine a Boston-to-New York City trip that takes about two hours—instead of the roughly 3.5 hours it can take today— for a fraction of the cost of prestige megaprojects.

The Wall Street Journal asked Levy to explain how small, coordinated changes can deliver big gains without the need for flashy tech, lots of new track or expensive tunnels, allowing U.S. rail to finally catch up to peers in Europe and Asia without breaking the bank.

Familiar, but better

WSJ: If your plans were implemented over the next decade, what would the passenger experience on the Northeast Corridor look like in 2035? ALON LEVY: It would feel familiar, but faster, cheaper and more reliable. You would board your train at a center-city station like Boston South Station, where your departure track is already printed on your ticket or app. No more staring at departure boards.

The trains would be modern, with smoother acceleration and better ride quality. You would travel from Boston to New York or New York to D.C. in under two hours. Fares would drop to about one-third to one-half of today’s prices for Amtrak’s Acela—say, $59 each way.

The real change is how normal long-distance train travel would feel. With faster trips and affordable prices, people would take these trains far more often—commuting, day-tripping, visiting friends, attending events. It would be normal to live in Philadelphia and work in New York, or vice versa. Trains would be timed to connect with commuter rail, making suburb-to-city travel seamless.

Overall, the biggest difference isn’t flashy tech—it’s usability. More people riding more often, without having to think much about it. That’s the shift.

WSJ: You proposed building near high-speed rail between Boston and Washington for $17 billion, less than a 10th of what some past plans have suggested. What are you doing differently, and why hasn’t this been done before? LEVY: What I’m doing differently is integrating every aspect of infrastructure planning with operations planning and timetable planning. This approach was developed in Switzerland in the ’80s and ’90s to avoid the high cost of building high-speed lines in a mountainous country too small to justify the expense. The Swiss trains aren’t especially fast—they’re just very well-timed.

WSJ: Your plan doesn’t rely on flashy tech like Maglev, a high-speed monorail technology that uses magnets to propel trains. Why stick with old-school steel-wheel trains? LEVY: First, other countries have stuck with steel wheels. It’s simpler technology, easier to build. You’re not locked into one vendor. With Maglev or similar, you need fully dedicated tracks. You can’t share infrastructure.

Using existing tracks for highspeed rail is crucial on the Northeast Corridor. Otherwise, you would need a new Gateway Tunnel between New Jersey and New York. The one being built already is expensive.

If we did go the flashy route, sure, you might cut the New Yorkto-Boston trip to one hour, 10 minutes, or New York-to-D.C. to around the same. But I’m proposing 1:56 using conventional rail. You can shave a few more minutes with expensive tweaks but the cost-benefit ratio just isn’t there. The marginal value of saving another 30 to 40 minutes drops off fast.

WSJ: You say most travel-time savings come from fixing slow zones, like the last mile of track into New York’s Grand Central. How does that work? LEVY: On the way into a station, multiple tracks come together with switches controlled by one tower that move each train to the desired platform, a process called interlocking. In the U.S., the approach is often set with very conservative speeds—for instance, a 10-mph limit into Grand Central— so trains crawl the last mile. Loosening these speeds where safe would save time without needing faster trains overall.

At 10 mph, it takes six minutes for the train to travel that last mile. But you could safely run most of it at 35 mph, maybe 25 mph near the platforms. That would cut the six minutes down to two.

More reliability

WSJ: How confident are you that better timetables could replace billions in infrastructure spending? LEVY: Very. We’ve run simulations repeatedly and found that in places like New York’s Penn Station, better timetabling could avoid $24 billion in expansion projects.

If you look at German scheduling, there’s a clear, regular pattern. A train starts around 4 a.m. and then runs every 20 minutes, all day. It’s predictable, stable.

In the U.S., that doesn’t exist.

Trains are individually scheduled and have different stopping patterns, meaning they stop at a different set of stations. That kind of variability is chaotic.

The result is schedules written with the assumption that trains won’t be on time anyway, so they’re padded and loose.

But if you do basic things like clock-face scheduling—say, every 20 minutes on the dot—it becomes much more stable.

WSJ: But if all trains on a commuter-rail line make—or skip— the same stops, doesn’t that reduce service for riders who rely on the variety? LEVY: Fewer patterns actually make service more reliable. If you have fewer train patterns, it’s easier to substitute one train for another if there’s a delay. So a 15minute delay doesn’t ripple through the entire network. That means trains can run closer together, which keeps frequency high.

WSJ: Won’t taking out express trains slow down service? LEVY: Not really. Right now, timetables are padded [meaning extra time built in for delays] by about 25%. If we simplify patterns, we can cut that to around 7%. Even if trains make more stops, they still arrive faster overall because there’s less built-in slack.

[Aaron Donovan, a spokesman for the Metropolitan Transportation Authority, says the extra time built into Metro-North schedules generally averages 10%, depending on destination and length of trips, and takes into account routine track maintenance and capital work that can increase runtime. Metro-North continually reviews models, signal timing, equipment, and other elements of operation to improve travel times and reliability for customers, he says.] [Amtrak says it is always evaluating its schedule and service and investing in infrastructure improvements to provide faster trip times. It says it already has several of the report’s suggestions in place.]

WSJ: How do you cut costs on rail projects like this without compromising safety? LEVY: Nearly all serious train accidents today come from overspeeding— trains entering a slow zone, like a tight curve, without slowing down enough. Modern systems like Positive Train Control (PTC) prevent this, so we can safely reduce padding in timetables without risking safety.

WSJ: Does electrification of trains enable the kind of cost savings you describe in your report? LEVY: Yes, absolutely. Electric trains perform so much better that they’re worth it almost everywhere U.S. trains run—just on return on investment.

Electric trains accelerate and brake faster, so local trains can stay closer to express trains. That reduces or even eliminates the need for passing tracks. Maintenance is easier. You need less yard space. Operations get simpler.

But electrification is underrated in the U.S., partly because of how poorly it was handled on Caltrain [the commuter rail line serving the San Francisco Peninsula that faced numerous delays and cost overruns].

Now, other agencies wrongly assume all electrification is that expensive.

WSJ: How easy will it be for politicians to sell a more modest approach like yours, instead of chasing prestige projects like we’ve seen in Japan or China? LEVY: At the end of the day, passengers don’t care how the nose of the train looks. They don’t care about top speed either, except maybe in brochures. What they care about is the timetable. “I leave Tokyo at 9 a.m. and get to Osaka at 11:25 a.m.”—that’s what matters.

If a flashy project delivers that, people will say, “Let’s do more flashy projects.” If a thousand small fixes deliver it, they’ll say, “Let’s do more of those.” What matters is performance and consistency.

If you look at the report, yes, it’s built on incremental improvements— but the results are transformative. Boston to New York in one hour 56 minutes. Same for Washington to New York. That’s a big leap from today—Boston to New York takes around 3.5 hours now, and the Acela to D.C. is about three hours, and more expensive.