Freakonomics Quorum: Can Amtrak Ever Be Profitable?

Amtrak’s ridership and revenue has been steadily increasing over the last 10 years, and 2011 set a new ridership record with 30.2 million passengers, and $1.9 billion in ticket revenue. But, even though it took in $1.42 billion from Congress last year, it still manages to lose $1 billion annually. This is hardly a new development. Amtrak has a long and storied history of functioning at a loss despite government subsidies.

So, as we enter what appears to be a new era (maybe?) of government austerity, it seems worth asking if Amtrak can ever turn a profit without government help. We rounded up some people who pay attention to this issue and asked for their ideas to fix Amtrak, if it can be fixed at all.

Thanks to everyone in the Quorum for participating — and as always, let us know what you think in the comments section.

Eric Morris, a regular contributor to Freakonomics, is a transportation scholar at UCLA. His work focuses on transportation and well-being, transportation equity, transportation finance, and transportation history.


I have a soft spot for the Zeppelin. This graceful ocean liner of the sky is quiet, clean, energy-efficient and—despite vicious Hindenburg-related media aspersions—safe. Whether providing us with spectacular aerial views of our favorite sporting events or bombing London, these delightful dirigibles have brought joy to millions. However, despite my personal love of the airship, I do not believe that government should be in the business of spending billions of dollars to support it when riders seem curiously indifferent to its merits. Supporting an unprofitable transportation mode because it’s “cool” is not a sound basis for public policy.

It is true that there are travelers out there who enjoy taking Amtrak. On rare occasions I have been one of them myself. However, society has no interest in paying for benefits—in terms of things like comfort, convenience or time savings—that are enjoyed by travelers. In our market system, which, recent speed bumps aside, is working quite well for us, the traveler himself should pay for the benefits he receives. If government has to step in and pay to produce a product customers won’t pay for, it is a sign that the product isn’t a particularly good one and is destroying wealth and resources that could be more profitably deployed elsewhere.

Thus subsidies for Amtrak are justified only if they bring broader benefits to society. Do they? Are oil rig workers in Texas or accounts payable clerks in Tacoma somehow better off because Joe Biden enjoys taking Amtrak back and forth between Washington and Wilmington?

The answer is almost certainly that any purported societal benefits from Amtrak do not match its subsidies, or even come close. Rail backers make a big deal about how much road and airport congestion their pet mode is relieving. But traffic congestion, while a big problem inside cities, is rarely an issue on highways between them. As far as airport congestion goes, in my experience I have never been troubled by a shortage of suitable flights. (Anyway, a better solution to airport congestion would be to modernize our air traffic control system, which at the present time is using technology not much more advanced than the abacus and the astrolabe.) Also, it’s not as if Amtrak is truly absorbing a lot of traffic from the other modes. In most of the country—even where it does exist—Amtrak’s ridership is about as significant as a rounding error. Even in the Northeast corridor, the one place where it is truly a meaningful mode, it accounts for only about 6 percent of intercity trips (driving accounts for 89 percent).

Are there any other benefits to non-riders that make subsidy worthwhile? Probably the only substantial one might be environmental, in terms of air pollution and greenhouse gases. It is true that moving things along rails requires less energy than moving them along roads or through the air. However, in most of the country Amtrak runs diesel locomotives which aren’t much better than internal combustion or jet propulsion. In the Northeast Corridor service is electrified, which is better. But even pro-Amtrak calculations (see this) find the pollution benefits are nowhere near Amtrak’s subsidy. And in fact, intercity buses are probably much cleaner than trains because the vehicles are lighter and generally travel filled closer to capacity.

Amtrak defenders like to make the argument that roads and air travel also receive subsidies. Sometimes it’s claimed that the total of these modes’ subsidies and externalities are equal to those of Amtrak. But my reading of the numbers is that this is questionable, and even if it were true it is a pretty weak argument. The fact that Fidel Castro is nicer than Kim Jong-Il is (or was) hardly a ringing endorsement for Cuban communism.

It isn’t hard to see what accounts for Amtrak’s failing to come anywhere close to covering its costs. Americans are not crying out in agony due to a lack of transportation options, and, simply put, passenger rail doesn’t have much to recommend it compared to the competition. On longer-distance trips, flight is vastly faster. The car isn’t much slower, is fairly inexpensive (particularly with multiple passengers, common on intercity trips), and provides far greater convenience, including door-to-door service and the use of a car when the traveler arrives at the destination. Intercity buses also aren’t much slower than Amtrak, are vastly cheaper, and aren’t that much less comfortable once they are on the highway. In terms of passenger transport, trains are a 19th century technology that we are attempting to apply to a 21st century problem.

Does this mean there is no room for rail in America? Far from it. Thanks to deregulation in 1976 and 1980, in the last few decades rail transport has been one of America’s great transportation success stories. However, this is when rail carries coal as opposed to commuters. Thanks to its characteristics (it is fairly slow and inconvenient but is energy-efficient and reasonably priced compared to trucks), rail is ideal for transporting many classes of freight, particularly cargoes that are heavy, low-value, and not particularly time-sensitive. Think bulk commodities like wheat, iron ore, or particularly coal. Passengers are light, high-value, and very time-sensitive, so it’s easy to see why rail isn’t right for them. Thus freight rail shipments have skyrocketed while passenger rail trips have stagnated. It is instructive that one of the first things the railways did after deregulation was dump unprofitable passenger service.

This is particularly important because freight and passenger rail transport do not go well together. Coordinating different types of trains traveling at considerably different speeds but sharing the same tracks is not particularly easy. This might be why Europe moves lots of people by train but the overwhelming share of its goods by truck, while in the U.S. the pattern is reversed. Perhaps it is impossible to make one rail infrastructure serve two master
s effectively.

Harsh as it sounds, my answer to the question of how to keep Amtrak from losing money is to eliminate subsidies and perhaps privatize the operation. Highly uneconomical service in most of the country will vanish, but frankly there are few riders who will notice and even these have plenty of other transportation choices. It is possible that in the Northeast Corridor train travel could have a fighting shot at standing on its own two feet in some form. But if it can’t, and we decide cars, buses, and planes aren’t serving the corridor adequately, I’d suggest we give up on trains, fold Amtrak entirely, and introduce legislation to found AmZep instead.

Robert Puentes is a senior fellow with the Brookings Institution Metropolitan Policy Program where he also directs the Program’s Metropolitan Infrastructure Initiative. He is an expert on transportation and infrastructure, urban planning, growth management, suburban issues and housing.

In Washington and across the nation, there are ongoing deliberations about which transportation and infrastructure assists will drive the next American economy. A particularly noisy debate involves the future of the nation’s passenger railroad network and where, in what form, and who should make these investments. These and other discussions have once-again raised questions about America’s national passenger rail system—Amtrak—which has faced a tumultuous future ever since its creation in 1971.

Despite the haranguing, Amtrak continues to enjoy support from many in Congress and is carrying more passengers than ever. In fact, it experienced a significant jump in national ridership after 1997 when a bipartisan federal commission was established to make recommendations to help Amtrak reach operational self-sufficiency. Since then, Amtrak’s total boardings and alightings have increased 34.9 percent. To put this in perspective, it more than doubles population growth (14.6 percent) over the same period and exceeds real GDP growth (29.5 percent).

Part of the reason is that Amtrak’s growth mirrors the rise of America’s largest metropolitan areas, many of which are served quite well by rail. In fact, half of Amtrak’s ridership comes from just five large metros: New York, Washington, Philadelphia, Chicago, and Los Angeles. These places are generally well positioned geographically with good connectivity to other key metros. They are also home to the nation’s largest aviation delays and highway congestion with which travelers in these metros have to contend. Indeed, Amtrak says it has a whopping 62 percent of the air/rail market between New York and Washington, and 47 percent of the market share between Boston and New York.

Why does Amtrak work in these places? For one, research suggests that for intercity rail corridors to be successful they require competitive travel times especially compared to air travel. Amtrak has a built-in benefit in some metros by providing direct service close to the heart of cities and business districts whereas air travelers must contend with decentralized airport locations, security lines, and early gate arrival requirements. And in terms of length, research based on European results finds that the optimal distances for travelers to shift from air to rail is around 300 miles or so. (Washington to New York is about 250 miles.)

It is these and other short-haul routes — some of which are aided by state financial support – that are the engines of Amtrak ridership. Indeed, when only considering corridors of 400 miles or less these short-haul routes consistently generate over 80 percent of all Amtrak ridership.

This interactive map by Pew Chartiable Trust’s Subsidy Scope provides an excellent illustration of how short routes that serve major metros perform better than other routes. A few, like the Acela and Northeast Corridor trains, produce greater profits than losses (including costs such as depreciation) while the opposite is true for super long haul routes like the Sunset Limited.

Of course, the question of government subsidies to all modes (highway, air, transit, bicycle) is an ongoing and complex debate.  But as it relates to Amtrak it does seem clear that focus and prioritization on short-haul corridors that connect our nation’s major metros is critical to its long-term success.

 Yonah Freemark is a journalist who writes about urbanism and transportation. He founded and continues to publish The Transport Politic. He has contributed to Next American City, The Atlantic Cities, Dissent, and Planning.

Compared to the intercity railroads of virtually every other developed country — and even those of many developing countries — Amtrak under-performs. Its services are simply too slow, too expensive, and too unreliable to attract a significant share of travel in the United States.

Yet the alternatives being hammered out by opponents of the nation’s railroad, such as privatization and a reduction in government subsidies, are a collective non-starter. Experience abroad suggests that successful rail systems – in fact, most transportation systems – require significant government aid, especially when it comes to capital projects.

Assuming that we believe there is a public interest in maintaining a functioning mobility network, then, there is no alternative but to continue federal funding for transportation. Railways offer the potential for uncongested service, are ecologically responsible, and are beneficial to the central cities they serve; the same cannot be said of rival automobile and aviation systems, which are plagued by delays, pollute massively, and encourage suburban sprawl. Train links must form an important element of America’s future multi-modal transport network.The question is thus not whether to cut off subsidies to U.S. railway operations entirely, but rather whether to continue support for Amtrak or provide aid to private companies that could take over its physical assets in the Northeast Corridor and its operations throughout the country, as has been suggested by House Republicans over the past few months. The latter option was pioneered by the United Kingdom in the mid-1990s, but the results were hardly laudatory as services became increasingly unreliable. The British government responded with the re-nationalization of that country’s passenger rail infrastructure in 2002, placing it back within the European mainstream.

U.S. intercity railway operations are underused and underserved compared to their British counterparts, due to decades of disinvestment and an overwhelming – and destructive – emphasis on automobility and aviation. There is little reason to suspect that it would be in the nation’s interest to privatize the ownershi
p of the railway infrastructure itself, since that has proven to be such a failure in similar conditions abroad. The American government, working with the states, must maintain ownership over the trackage.

But how about operations? In the U.K., passenger services are now provided by private companies. Yet on many corridors, operations are heavily subsidized by the public sector, and even some of the supposedly profitable lines have been abandoned by private operators, only to be served by a publicly owned holding company. Overall subsidies to passenger rail in Britain are far higher now than they were when the railways were run by the nationalized company, British Rail. It seems unlikely that the far less-used American intercity railway network — outside of the most-frequented segments of the Northeast Corridor — could be profitable for private enterprise; while subsidies to Amtrak may be unpalatable, subsidies to less accountable private companies seem even less desirable.Thus an improving American railway network requires a functioning Amtrak.

None of this is to suggest that Amtrak should not strive towards increasing speeds, decreasing fares, and improving reliability – far from it. The Obama Administration’s investments in intercity rail will bring faster and more dependable Amtrak services to states like Illinois, Michigan, and North Carolina within the next five years. Similar improvements over the past decade have led to dramatic increases in the company’s ridership along such routes as California’s Capitol Corridor and Pennsylvania’s Keystone Line. In addition, the national railway company must invest in increasing its capacity, a step it is already taking with new cars for regional services and its Acela trains on the Northeast Corridor. More room on trains will mean increasing efficiencies of scale, potentially lowering ticket prices for consumers. Finally, Amtrak must make an effort to reduce capital construction and operations costs, both of which are higher than those of similar railway companies overseas. Without this step, the country will continue to be saddled by railway operations that cost too much for an already overburdened government. The road ahead for American railroading will continue to be paved by subsidies. But the strong train system that will result if properly planned will be beneficial to riders and the nation as a whole.

Nate Berg is a journalist covering cities, the environment and urban planning. He has written for a variety of publications, including The New York Times, Wired, the LA Weekly, and many others. He is a staff writer at The Atlantic Cities blog.

In 2002, five years after Congress asked Amtrak to find a way to not require federal operating funds, David Gunn, five weeks fresh as Amtrak’s new president, flatly told a Senate committee what everybody already knew: “Amtrak will never be profitable.”

The passenger rail service, needless to say, did not meet its mandated goal of breaking free from government subsidies. This shouldn’t be a surprise. The vast majority of the world’s public transit systems rely on government subsidies. Though there are some passenger rail lines in Europe and Asia that can operate at a net profit largely without ongoing government support, train lines are more often government services than private businesses. To call on Amtrak to make money ignores the fact that it was created and still today operates as an infrastructural service. The real issue is not so much that Amtrak will never be profitable, but rather that it shouldn’t have to be.

Amtrak is infrastructure in the truest sense of the word. It’s one of the physical structures and services that organizationally enable the functioning of U.S. society. Like the interstate highway system or the postal service, Amtrak provides a service that (to an arguably lesser degree) serves and benefits residents and businesses. These services may not have direct dollar returns, but to say they don’t fill a vital need would be ignorant. Calling on Amtrak to go it alone prejudiciously holds the service to a different standard than other government services.

The return on investment for these services doesn’t have to be direct, and for transportation modes like passenger rail it often isn’t. Think of transportation as a line connecting two dots. The economic value of this connection is captured not by the transportation method itself, but by the people and places it connects. The value is in the dots, not the line. But without the line, the two (and in reality many) dots would have trouble sharing and compounding their relative economic powers.

These dots, the mid and endpoints of these transportation corridors, though, are not just single businesses or factories. They are cities, which are increasingly the centers and enablers of economic power. And while it’s true that Amtrak lines aren’t the only ways to get from one city to the next, they can play a valuable role in allowing people and services to travel between these economic centers in as many ways as possible.

In their new book Megapolitan America, Arthur C. Nelson and Robert E. Lang note that about 65 percent of the U.S. population lives within 23 clusters of metropolitan areas. This level of concentration shows the significance of cities, but also their interrelatedness. As groups of cities and their suburbs grow more populous, the physical transportation connections between them – highways, airports, Amtrak lines – become even more important. Rather than being seen as separate, Amtrak should be considered a part of the transportation network the government has enabled to connect these places.

Certainly Amtrak could be operated in a way that better met the needs of our increasingly mobile and interconnected society. But to single it out as a money hog amongst a broad swath of government-subsidized infrastructures and services is to turn a blind eye to the fact that the provision of such services, profit or no, allows the country’s economy to function.

Stephen Smith writes for Forbes on the politics, economics, and history of urbanism. He formerly wrote for the Market Urbanism blog, Reason, and the National Review.

In many ways, Amtrak’s problems are bigger than itself. They have their roots in one of America’s most enduring trends, which is population dispersal – whether it’s through westward state-funded canals and railroads of the early days of the republic, or twentieth century automobile-based suburban and exurban sprawl. For transit of any kind to be truly revived, land use patterns will have to be significantly altered, along with some of the most important tenets of American culture.

That said, the United States is taking tentative steps towards re-urbanization, and intercity travel is generally the most profitable kind of passenger rail (perhaps because there isn’t as much political pressure to keep cap fares). The consensus in the European Union and Japan is clearly in the direction of privatization. Even the privatization-averse Swiss Federal Railways aims for profits on its main intercity lines, while achieving remarkable gains in ridership over the past decade. Liberalization seems to be the future of intercity rail (to say nothing of the past), and there is no good reason that American politicians shouldn’t begin the process towards privatization now.

The process, however, will be a long one, with privatization being the end goal rather than the starting point. The first thing that must be done is not even management-related, but rather regulatory: the Federal Railroad Administration’s approach to rail safety must be radically restructured. The FRA can no longer be allowed to lag behind
European and Asia regulators, foisting unrealistic and outdated safety mandates on all mainline passenger rolling stock in the US, from Amtrak trains to regional/commuter railroads. Our unique standards for bulk and other more technical aspects of rail car design should have been abandoned half a century ago.

America’s FRA-compliant obese mainline trains are slower, more expensive, and less reliable than modern European and Asian designs. Getting this right is the most important step in the whole process, partly because it will help “commuter” services like the Long Island Railroad and Caltrain. Good intracity mass transit links are a key to good intercity rail, and FRA regulations are more onerous for commuter railroads than they are for Amtrak.

Next, management must be given much more latitude when it comes to labor issues. They should be able and willing to negotiate (and risks strikes, if need be) over everything except wages and benefits, with the minor proviso that pensions may need to be reigned in. Managers must, however, be given control over work rules. A century ago, when railroading was a grueling and deadly profession, work rules were an essential negotiating point. But now they’re lucrative sources of indirect compensation, and ripe for featherbedding. There may come a point in the future when salaries and benefits should be negotiated down as well, but currently it’s a minor issue compared to work rule reform, and is not worth the political will it will take to tackle.

Once management is empowered to take control of the organization, we should expect a better financial position, and it will finally be time to begin thinking about privatization. The reason deregulation must come first is that if we were to privatize Amtrak first, as Rep. John Mica‘s plan earlier this year would have had us do, the privatized firm would just face the same difficulties that Amtrak’s predecessors faced before nationalization in 1971.

But even privatization is not a straightforward affair. The first decision to make is whether to go for the European model of forcibly separated infrastructure and operations, or to sell off the railroads as an integral unit, which is the path that Japan and the United States have historically taken. Unless the E.U.’s open access policies are wildly successful by the time deregulation is finished, it would probably be best to stick with the integral model.

But beyond that, we have to decide how to break up Amtrak. The Northeast Corridor is an obvious candidate for splitting off and selling, because it’s both the most profitable route, and the only one whose tracks Amtrak actually owns. The rest of the network will almost certainly need subsidies to survive, although popular routes like those in California should be further separated from inland services, so that in case they are seen as profitable in the future, they won’t be held back by the morbid routes away from the coasts and major cities. To what extent America outside of the Northeast gets subsidies is a political question, but it should be strictly segregated from the viable Northeast Corridor.


I hate that we as Americans are obsessed with the idea of "profitability". Is there no other way we can think? There are a lot of things in life that aren't profitable, but are important for society as a whole. Maybe Amtrak is one of them. People who prefer to drive cars might not like having their money go towards a service they never use, but how can they not appreciate the fact that trains running between major cities take cars off the roads they do use, leading to less congestion? In the Northeast at least, this is as close to a no-brainer as can possibly exist.


I'm a little surprised they operate at a loss. Every time I've taken Amtrak the train has been pretty close to full for most of the trip, and the ratio of workers to passengers doesn't seem excessive - at least not compared to air travel. I prefer to fly, but I more prefer to skip the TSA hassle whenever possible and take AMTRAK for long trips.

Eric M. Jones.

Eric Morris writes: "I have a soft spot for the Zeppelin. This graceful ocean liner of the sky is quiet, clean, energy-efficient and—despite vicious Hindenburg-related media aspersions—safe. Whether providing us with spectacular aerial views of our favorite sporting events or bombing London, these delightful dirigibles have brought joy to millions."

A transportation scholar should be clear on definitions. Zeppelin is German for dirigible (rigid airship), and they are are not blimps (non-rigid airships). Dirigibles are much faster than blimps, and much more fragile. The history of dirigibles is a nightmare; not just the Hindenburg. The average life of a dirigible could be measured in single-digit years, usually ending in disaster. They were also horrendously expensive to build and operate. A large dirigible cost as much as a B757, and its operating cost was much higher.

James A.

Mr. Morris writes, "Rail backers make a big deal about how much road and airport congestion their pet mode is relieving. But traffic congestion, while a big problem inside cities, is rarely an issue on highways between them."

Assuming for the moment that this point were true (which I do not concede), I would like to point out that a trip by car between two cities generally involves driving within both of them. Or would Mr. Morris suggest that a trip from Washington to New York would not be affected by traffic congestion in either city?

Further, he writes, "As far as airport congestion goes, in my experience I have never been troubled by a shortage of suitable flights."

This quote suggests a fundamental lack of understanding of the term "congestion". Congestion is a condition characterized by a decrease in performance due to infrastructure being utilized beyond its functional capacity. It has nothing to do with the availability of vehicle trips (or lack thereof) or those vehicles' capacity.

That a supposed "transportation scholar" would make statements as wildly ignorant of fundamental transportation principles such as these does not speak well of his credentials or those of the institution he represents. And I say that as a transportation professional with an M.S. in transportation systems engineering.


Enter your name...

One stray thought I have after reading all the comments:

Does anyone else remember how relieved we were on 9/12/2001 to discover that America still had a passenger rail service? With air travel completely shut down, many thousands of stranded travelers got home on Amtrak.

In addition to its practical value for shorter trips and for a few people who cannot fly at all for medical or legal reasons, keeping Amtrak around is a bit like making a backup of your computer hard drive. It's wise to have an alternative in case your normal plan fails.


Seems to me that having a completely second transportation system that almost no one uses in general is a bit overkill for a once in a, what, century? two?, scenario. Like having a second, always up to date, just as expensive but not as good computer in your closet in case your first one crashes instead of an inexpensive backup solution.

James Harris

All I can say as a British citizen is whatever you do don't do what the Brits did. Our privatized railways became a basketcase for a decade and despite current record ridership still currently display a huge variation in quality and frequency of service, with trains running half-empty at expensive prices being considered better than full trains on the cheap. Some things are best run for profit, and some for the public good, and for me railways belong in the second category. The Amtrak substidy is measly, compared to for example the US' 700 billion dollar defense budget, and for me represents a good and growing return. I currently live in Germany where thousands U.S. troops are still stationed. If you bring those troops home, that might free up a little money to afford such necessities as a decent train network.


The car-driving oil roughneck in Texas may not be happy about paying for the rail I use.

But I and the other million or so northeasterners who DON'T drive aren't too happy about paying for the highways he uses. If all highways become toll roads and Congress ends highway subsidies, so that road users bear the burden of road use, then people might see rail as more of a reasonable substitute for their travel.

Liz B

Puentes noted the environmental public benefits of moving travelers from cars to trains, but I'd also like to see more discussion of passengers for whom driving or flying is not a good option.

Car ownership and use is expensive, and will only become more so as fuel prices keep rising. Many people don't own cars at all, and intercity bus service isn't always available for them. (Greyhound cut its service substantially in the mid-2000s, and the recent growth of inter-city bus travel seems to be mainly between large cities.) Flying can be expensive, especially for those who have to travel long distances to airports that are inconveniently located and/or poorly served by public transportation -- and dragging a suitcase onto multiple bus or train lines isn't feasible for everyone.

Rental cars are an option for some people who don't own cars (my non-car-owning household loves Zipcar for weekend trips out of the city), but not everyone can drive a car safely. Those who have slow reaction times or poor vision or who are impaired by alcohol, fatigue, or distractions represent a danger to themselves and others when they get behind the wheel.

To some extent, rail and intercity bus are interchangeable for medium-distance trips, although rail has the benefit of neither influencing nor suffering from car-traffic congestion.
In order for society's poorest and least physically capable members to travel outside of the cities and towns where they live (a reasonable requirement for a developed country), a strong network of intercity bus and/or rail is essential.



As much as I would like to, I can't justify using the train to go from Peoria to Chicago. For what takes 5 hours and 3/4 a tank of gas (round trip), I'd have to take a bus into the city center of Peoria, another bus to Blomington (an hour away), a train into Chicago (1.5 hours), local transit to where I want to go (about an hour), and reverse to get back. 7 hours, two buses, two trains, and no transport or freedom to be on my own schedule for more money than the 3 hours and 12 gallons of fuel a car costs me.

I'd love it if there were trains that followed all the major interstates, especially away from major cities. Peoria's a pretty big town, but there's no train through the city and no trains, as far as I know, that run east to west around here. If there was one traveling along I-74 that I could hop on, it'd be way more convinient.

Lou Thompson

The articles are curiously unaware of what Amtrak actually is and why it is that way. Amtrak was actually created because the private freight railroads, who had been required by the ICC to carry passengers at a loss, could no longer carry the financial burden. The idea that Amtrak would ever be profitable was part of a political narrative at the time that was about as credible as the current narrative that we can solve our deficit problems without raising taxes. In fact, shifting the burden of passenger losses to the public sector, where it should be, played a major role (subsequent deregulation was more important) in recovery of the profitable and efficient freight railroads that the country enjoys. In other words, Amtrak was definitely a success if you understand the actual objective. The subsidiary objective -- saving passenger service -- not so much.

It is misleading to talk about "Amtrak" as an entity: instead, Amtrak is actually a grouping of three, quite different functions -- the long haul trains, the short haul trains and the Northeast Corridor.

The long haul trains do not perform much of a transportation function: they are instead an option that Congress has demanded that Amtrak perform. They are, incidentally, not infrastructure as they only involve Amtrak as a tenant on the lines of the private freight railroads. All of the long haul trains lose money (revenue-operating costs) and many do not even cover their labor costs from revenues. Of course they could be operated by the private sector as "negative" concessions (maybe some could survive as tourist cruisers).

The short haul trains either have (or may have) a real transportation role into and out of congested urban areas, but they are also primarily local rather than national in scope. It really makes no more sense to have a single, national company operating them than it would to have a single, national operator combining BART, CTA, MTA and MBTA. Congress and the DOT are increasingly and correctly forcing the states to take a larger role in planning and financing these services and the states (equally correctly) are asking for a larger role in control over operations, even extending to concessioning. In almost all the short haul routes, Amtrak is a tenant on lines operated and maintained by others. Like the long hauls, the short hauls require operating support (some more than others).

The Northeast Corridor is quite different, partly because of its clear purpose as an important transportation option in the Northeastern megalopolis and partly because Amtrak is not a tenant, but owns the infrastructure for most of the way from DC to Boston. There is only limited operational and service interaction among the three components of the company called "Amtrak." Amtrak's Acela covers its operating costs with a significant margin and the regional trains do not fall far short. In addition, the NEC tracks carry a large number of commuter trains for which the access charges are probably below marginal costs and freight trains with access charges that are probably above marginal cost. In short, there is a good argument to be made for developing the NEC, but with a larger regional role in planning and perhaps financing although the Federal Government actually owns the properties that Amtrak manages.

The real problem with Amtrak, as with many current transportation programs, is that it is actually a mix of public and private benefits and costs. In Amtrak's case, we already know that the private benefits do not cover the private costs: we learned that well before 1970 and have only continued to prove it since. In the U.S. unlike Europe, we are not good at publicly identifying and reliably paying for public benefits (we prefer to force the private sector to pay if we can or, increasingly, we seem to prefer not to pay for them at all). The private sector will operate and pay for freight services so long as we don't mis-regulate the industry (ever a temptation), but the private sector will not, and should not be asked to provide rail passenger services unless the public pays them to do so.

Can the private sector provide public services effectively? Of course they can if the terms of the public/private partnership are clear and balanced. The UK experience is not an unalloyed success; but, then, neither was the predecessor British Railways (BR). Yes, costs are up in the UK, but BR had been under-financed for years and the piper had to be paid. In addition, one reason why public investment is up is that traffic began growing immediately after franchising and traffic levels are well above the historical high at the end of WWII. UK rail passenger traffic has also grown faster after franchising than any other railway in Europe. Other EU governments are experimenting with franchising with enough success that re-nationalization of operations (as opposed to infrastructure) is not on the table anywhere.

Franchising (or concessioning) is not simple and there have been a number of well documented problems. Then again, compared to what?