Amtrak: Losing Money for the Long Haul

Why can’t Amtrak make any money? America’s national passenger rail service is a for-profit corporation, yet since its trains started rolling in 1971, Amtrak has yet to come out in the black.

By generous accounting measures, Amtrak’s operating losses in fiscal year 2015 were $306.5 million. It survives on government subsidies, but with these subsidies come operational mandates that ensure Amtrak stays unprofitable.

Amtrak’s business model is to transport rail passengers in the United States with objectives of safety and security, customer focus, and financial excellence. But a dissonant operational model drags it down on all three fronts.

Covering the United States by Passenger Rail, No Matter the Cost

Amtrak operates trains with destinations in 46 states and three Canadian provinces. Compared with the strong intercity rail systems of Europe and East Asia, it covers areas of lower density and further flung cities. This is part of the bargain behind Amtrak: in exchange for its monopoly on passenger rail, it has to keep long-haul trains running all across the country.

AMtrak Map


In terms of costs, Amtrak’s long haul trains are horrendously inefficient. Consider the “Empire Builder”, which runs from Chicago to Seattle, with plenty of seats empty as it runs through northern Montana and North Dakota. The route runs at an annual operating loss of over $10 million peramtrak year, as does the “Southwest Chief” from Chicago to Los Angeles, the “Sunset Limited” from New Orleans to Los Angeles, and the “California Zephyr” from Chicago to San Francisco, which loses over $30 million annually. Altogether, long distance routes account for about 15% of Amtrak passengers – and 41% of costs.


Amtrak makes much more financial sense across short high-density routes. It is particularly popular in the Northeast Corridor; trains accounts for 77% of all air and rail travel between Washington and New York, a share that has been growing since the introduction of Acela Express service in 2000. Overall Northeast Corridor operating profits exceed $200 million annually with ridership of 750,000 people per day.


Safety and Security in Question as Capital Erodes

Amtrak’s has put together a fairly strong long-term record of passenger safety. This past May, concerns regarding its ability to maintain this record were highlighted as a train derailed outside Philadelphia, killing seven. The crash was attributed to driver error, but it could have been prevented by modern systems of signaling and automation. Experts point to the chronic underinvestment; the train fleet is currently as old as it has ever been, and rail lines suffer from widespread disrepair. The operating surpluses that might have been poured into infrastructure investment in the philly-crash-mondayNortheast Corridor are instead used to sustain money losing routes. Necessary and capital intensive updates to the 1873 Baltimore and Potomac Tunnel and the addition of a Hudson River tunnel have been long delayed as they will depend on the sort of federal funding that is difficult to sustain over a long project.


Amtrak cited cleanup, repair, settlements, and downtime resulting from the Philadelphia crash as the primary reason its operating losses yawned wider in FY2015. A vicious cycle seems to be emerging: the expensive results of safety failures leech funding away from infrastructure investment, leaving Amtrak ever more vulnerable to further accidents.

Customer Focus Lost Among Big Expenses

Amtrak’s inability to invest in infrastructure also limits its ability to fulfil its business model’s tenet of customer service. While trains between Barcelona and Madrid average 154 MPH, the “high speed” Acela train carries passengers at an average speed of just 68 MPH. Under the current operating model, “real” high speed rail is at best decades away in the Northeast because of the extensive and expensive track overhaul it would require. Meanwhile, smaller and more incremental improvements are also on hold. Customers complain of shoddy wi-fi service. Food services are often cut back as their expense serves as an easy, if small, target for Congress.

8599657090_96e9774cc8_bThe Baltimore and Potomac Tunnel

Can Amtrak move forward?

If Amtrak operated as a true private business, it would first grasp at solvency with an obvious operational change: shutting down long haul lines. Instead, it has taken half-measures to plug funding gaps. The Passenger Rail Investment and Improvement Act of 2008 required states to contribute funding to support routes with local stops, and funding deficits quickly shrank, but the requirement was limited to lines of less than 750 miles and thus excluded the system’s most heavily subsidized routes. More recently, a federal transportation spending package signed December 7 stipulated the launch of a pilot program in which private carriers take over operation of some long haul routes. Given these routes’ losing propositions, though, it’s hard to imagine any sane company stepping in and upending the status quo. As long as Washington keeps separate Amtrak’s business and operating mandates, and as long as it keeps each line’s function separate from its profits, American passenger rail will keep chugging along, slowly and in the red.


Allen Edmonds: An American Revolution


Chipotle: The birth of the fast-casual concept

Student comments on Amtrak: Losing Money for the Long Haul

  1. Will– this is spot on. I took the California Zephyr from Emeryville to Chicago over the summer, and the scheduling is terrible. We arrived 10 hours late to Salt Lake City, and the Amtrak employees thought this was normal. I learned that Amtrak often sells its passenger rail right of way to trains that are shipping cargo in order to boost revenue. Cargo rail is much more profitable than passenger rail for long haul routes in the U.S.

    Interestingly, the new Hyperloop system is taking a cargo-first approach as well:

  2. Will, this is a really interesting article. It would be interesting to know if Amtrak has looked into other areas of revenue generation beyond simply serving passengers. In Alaska we would often face a similar issue with airplanes, in that our airplanes would never achieve the scale necessary to make a run profitable. So what wound up happening is that the airplanes would take their extra capacity and use it for cargo/freight service. Alaska Airlines was able to get contracts to deliver mail for the postal service on their planes, which helped to make the runs more profitable. It would be interesting to see if Amtrak could do something similar, and use their excess capacity to deliver certain forms of cargo as a way of closing some of their profitability gap. You would think that if the government were already propping them up, we could at least be using their network as a way to transport something like mail or parcel service, perhaps at a lower rate than what we could get from air cargo.

    My wife an I love the idea of Amtrak, but have always wound up passing on it because it winds up being so much more expensive than other forms of travel. If they were able to get their costs and prices under control, it might actually be a concept that would make sense. I like your ideas for how the company could improve!

  3. This is fascinating, Will. I wonder if this vicious downward spiral of Amtrak’s is emblematic of broader infrastructure issues we are facing in the US. Is it possible that people’s travel preferences are changing, and there simply is no need for connected rail travel? I have to wonder if the fact that they make most of their money off of the Northeast Corridor route is the market’s way of telling them it’s time to shut down. If they did close, what would they do with all of the spare parts?

    Alternatively, could Amtrak possibly pivot to focus their business model on something else? Similar to what Sam suggests, could they change their pricing structure to be more appealing to customers on less desirable routes? Could they partner with trip organizers to make Amtrak part of a whole package instead of a means to an end?

  4. I was drawn to your article Will as I was surprised coming form Europe at how terrible the rail system is in the US. I wrongly assumed the issue was a lack of federal funding and that Amtrak was trading off bad service and high prices for profitability!
    As you mention, the Montreal to Washington axis would be a great market to develop. Comparing French rail costs and speed, a Boston to Washington D.C. trip would take about 3 hours and cost around $200 return (same distance as a Paris-Marseille). The Acela takes over twice that time at twice the cost! A faster train could beat air travel not only in terms of time and cost but also in terms of convenience and labor productivity. You can be much more productive when spending 3 hours sitting in a train than when you are interrupted every 30 mins or so (20 mins cab ride followed by 20 mins at security check, 20 mins for boarding, 80min in the plane, 10 mins waiting for your check in luggage, 20 mins uber to get back to the city center!). This would kill the plane for these short hauls routes like it did in Europe which incidentally would also be much better for the environment as trains emit much less pollution than planes…

    I would however push back on the libertarian solution of a privatization of Amtrak. As Willy briefly mentioned in the presentation he gave over lunch last week, large “common good” investments such as infrastructure are rarely done by private businesses. That’s because they think short term (if you are lucky medium term) profit and ignore the byproduct effects of their investments. The long term economic repercussions an efficient high speed railroad would have are much wider than the profits it would generate for Amtrak. No doubt government management of Amtrak has been dismal. However, I would argue the solution could be for them to redefine the mission Amtrak aims to deliver and get their priorities straight rather than full on privatization. As Ellen points out, what is the value add of a passenger Chicago to L.A. train when you can fly? But I realize this is more a BGIE debate than TOM…

  5. Ah Will! You’ve found every East Coaster’s biggest gripe with this one! It’s really a shame how inefficient the system is. Even in the Northeast corridor, as Laurent points out, the trains could be so much faster. My understanding is that the Acela trains themselves are not the limiting factor, but rather the tracks and overhead lines cannot support high speeds all along the route. They also run into problems with sharing spaces with freight and commuter rail. The investment to fix these kinds of infrastructure problems would be immense, involve other public and privately run rail entities, and would like require cutting service while upgrades were performed. As a semi-public entity, major change would be very hard for Amtrak to accomplish. I give them some credit for small improvements overtime, including the app, delay alerts, and the ability to change your ticket on the run from your mobile device rather than waiting at the ticket counter or kiosk. While it inconveniences the traveler, no longer offering refunds for missed trains was probably a smart financial decision. Many of these improvements are far from smooth, but they do utilize technology to capture some low-hanging-fruit.

    As you note, the business model and operating model here are two separate things and the organization does not have the authority to unite them. It’s hard to see a scenario where the tough decisions Amtrak would need to make to become a well run entity would make it successfully through the legislative process. There is certainly a question of whether this business would be better managed by the private sector, but in that case, would we be left with any rail transit at all?

    1. Ashley! Great question! Reading Will’s post and your comment reminded me of the fact that both public and private rail can and do exist, but not in the US.

      I traveled a lot in Europe this summer and found that part of what makes the EuroRail system work so well is the mix of public and private companies providing train service. My typical experience in Germany was choosing between a slower, less direct route typically run by the government or a more direct train run by a private company.

      A key to this side-by-side model is infrastructure (rails, train cars, etc) that are functioning well. Things are on time and stay on time. The slow and the fast trains can live on the same system as long as each arrival and departure happens as scheduled. The lack of density along train routes in the US increases the cost of upkeep and might be one additional barrier to realizing public and private rail companies on the same tracks.

  6. Will, great post! Having lived in Spain for a year in high school where the fast trains indeed were fast, I’m perplexed people pay so much more in the US to take the Acela, which hardly saves you 30 minutes on the Boston to New York routes. I used to take Amtrak a lot between Boston and New Haven to visit my husband, and there is one hidden gem about Amtrak which is that their customer service over the phone is (surprisingly) excellent. I often had to change train tickets, and it was always faster to call in than log in online because the customer service agents were so efficient. Not to mention they often gave me great tips for how to get better deals! Alas, that’s clearly not enough to make up for their poor results and slow service. Getting to New York in 1.5 hours instead of 4 on the train (without the security hassle) would be game-changing and if they could achieve these results through additional funding or savings from shutting down some of the less profitable routes. I remember in Spain, getting all the provinces on board for the first high speed trains took a lot of time and so I imagine the same issue applies in the US. Do you think we might be more successful at launching high-speed rail within one state, for example between San Francisco and Los Angeles, rather than across state lines?

  7. To echo Ashley — you’ve certainly pinpointed a major pain point here for East Coasters! Every Thursday night I used to take the Acela Express back to Boston from Newark, NJ, a trip that in the best of times took 4.5 hours but which regularly took 6-7, especially in the winter. Though flying was an option, I was always drawn to the ostensibly more efficient train system (walk on, no security or baggage limit, no lost time waiting in ticketing lines, etc) — and was invariably disappointed. Like Annie, I’m excited to see what the Hyperloop has to offer, but have little faith that we’ll see a functional AND profitable system like that in the US in our lifetimes.

    As a sidenote, I’ve taken high speed trains (like this one a bunch and they are amazing! Pretty cool to see what technology and infrastructure can do in other places, though there are clearly some big issues with the way the Chinese gov’t went about building these.

  8. Nice article, Will. It’s interesting to see the US/EU comparisons that have been drawn out with the public good vs. privatization argument. Another consideration that comes to mind is the extent to which train service currently fills the void in the US between really cheap, low quality bus transport and higher priced, more efficient plane transport. It works for the Northeast Corridor, where this niche is profitable, but the trade-off is only a couple of hours vs. ~$100 and people are willing to accept lower quality wifi and food services. If an upgrade in services and speed occurs, and there is a corresponding increase in price, I don’t see this new train service being differentiated from a plane trip.

    To your other point, I see the customer’s acceptable time/money tradeoff as limited to regional routes only. For Chicago to LA, even a bullet train at 200 mph would take 10 hours (vs. 4.5 hours flying) – so even bullet trains likely do not have a place for longer routes. I agree with your assertion that even privatization of these longer haul routes may not make the operating model any more sustainable. A more interesting model would be an integrated model of plane and train travel (where planes handle the longer more profitable routes, and trains handle the shorter regional routes). However, they would face a challenge in facing off against smaller, regional airlines (e.g., Southwest).

  9. Love the write up Will and I thought I’d hop on the hate train and leave a comment. Its unbelievable that Amtrak is able to do what it does. What a waste of the taxpayers money on an outdated technology that lags behind every mode of transportation today. I see it as a classic case of corruption in the US government, where Congressmen are too afraid to put a stop to a program that loses money so long as it provides jobs in a district. I think it will be interesting to see how the two companies working on developing the Hyperloop will affect the US rail system in the coming years. Hopefully, they put Amtrak in the grave.

  10. Will, very interesting article. While I certainly agree that the service and especially the on-time performance Amtrak offers is far from being acceptable, my question is, if a national rail company has to be profitable. When looking at the railway history, the railway was never only a means to its end. Regardless whether we look at Europe or the US, the railway has always been closely connected to personal freedom. At its introduction it was the first time that middle class people could afford to travel reasonably fast and save. As railways connected cities and people and enabled trade and business, government historically always supported the development and maintenance of the infrastructure and also covered operating losses. The believe was that the railway created a greater good for the society that far exceeds the government spending on infrastructure and operating expenses. Therefore the railway was also an instrument to reduce social disparity and enable people to travel that cannot afford a car or a plane ticket.
    It’s true, many things have changed since the railway became big; still the questions remains, is the railway purely a means to its end (earning money by bringing people from A to B) or contributes a functioning railway system to a better society (enabling people to travel that cannot afford to travel otherwise and supporting business and trade).

Leave a comment