Revisiting Privatization in Intercity Rail

» Amtrak, as always, is being targeted for privatization by conservatives. But what approach leads to optimized public benefit?

Over the past few weeks, U.S. House Transportation and Infrastructure Committee Chairman John Mica (R-FL) has convened a series of hearings on the failures of Amtrak, America’s independent — but fully federally owned — national rail operator. Mr. Mica has used the meetings to wage an ideological crusade against the railway, arguing that it is too inefficient and expensive to continue receiving subsidies. Republican Presidential nominee Mitt Romney has also advocated selling Amtrak.

Democrats have mostly shot back, arguing that privatizing the agency would result in a significant reduction in services provided and increase ticket costs.

Here is the confusing truth about Amtrak, however: The rail agency, fully government-owned, is in many ways already a privatized operation that receives federal subsidies. The organization does not seem to have the larger public’s interests in mind in setting policies: It has some of the highest fares in the world for services in the Northeast Corridor, it provides no discounts for people of lesser means, and it actively promotes the use of intercity buses for people who want to pay less, in effect strategically reducing its market share. These are not the actions of a government enterprise acting in the public interest; these are the actions of a private corporation attempting to maximize profit.

Amtrak is an agency that, in its existing form, can satisfy neither the left nor the right.

And of course, Amtrak is not making a profit — it operates at a considerable deficit every year. But “privatization” can mean many things — and is a private alternative any better than the current circumstances?

Over the past month, Stephen Smith has published articles on the failure of Britain’s rail privatization, the success of Japan’s effort, and the potential for privatization of Amtrak services. The overall conclusion from the articles seems to be that there may be benefits from the movement of rail services out of the public sector, but that without vertical integration — in other words, single-company control of both tracks and services — difficulties are likely to ensue, as in the case of the U.K. Smith points to the structure of Japanese National Railways (JNR), in which six private companies control services in different parts of the country.*

Smith suggests that with privatization, Amtrak could radically improve its efficiency. The biggest problems with the rail agency, he argues, are related to low worker productivity. Despite Amtrak’s privately motivated interests that I pointed out above, much of its labor rules remain affected by politics and can be altered by Congressional action. Are we willing to accept reducing the influence of democratic actors in agency decision-making? It would mean restructuring labor agreements — reducing the income and health benefits that unions have fought for decades to acquire — and firing huge numbers of workers (a third in the case of JNR’s privatization).

If privatization slashes the number of workers needed to do the same job, rail in the U.S. could indeed become profitable, since most of Amtrak’s costs are labor related. For the transportation public, that could produce cheaper ticket prices and fewer subsidies. But it means, fundamentally, that we are bringing private companies in to do the dirty work that the government is politically incapable of doing.

Assuming that some privatization is accepted as “needed” to improve American rail service, in what form should it be implemented?

In general, there are four rough frameworks for such privatization:

1. Publicly owned tracks with competition for services. This is being implemented in mainland European countries under E.U. regulations; public sector track owners (such as RFF in France) allow operators — both public and private — to run competing services on the same lines.** This allows riders to choose operators on journeys with the same origins and destinations, just as can be done for airline journeys.
2. Publicly owned tracks with competition for contracts. This is the network organization in the United Kingdom; public Network Rail owns the tracks but then leases the rights to operating rail corridors to private companies. In general, contracts last around seven years and give each operator close to monopoly rights over each corridor.***
3. Privately owned tracks with competition for contracts. This was the system previously operated in the U.K.; the privately controlled Railtrack owned all tracks in the country between 1994 and 2002. The tracks were moved into the control of a public operator, as described in the second alternative.
4. Privately owned tracks with one private operator. This is how intercity rail operations are managed in Japan.

Smith and other critics have argued that separating track ownership from rail operations — that’s what happens in alternatives 1-3 described above — result in inefficiencies and potential competing motivations. The U.K.’s attempt to privatize both tracks and operations (alternative 3) produced a number of failures, reducing the safety of the services provided. These difficulties led to the nationalization of the track ownership (a switch to alternative 2), but the contractual relationships between the track owner and the rail operators continue to be a cause for concern. In both cases, the U.K. government has subsidized capital upgrades and in some cases it has subsidies operations.

The problems have expanded over time. Operators have failed to follow through on their commitments. In some instances they have promised too much. In the case of services on the East Coast Main Line, operator National Express East Coast gave up its contract following higher costs and lower revenues than expected; the result was that the government took over operations directly through publicly owned (but supposedly temporary) Directly Operated Railways. National Express’ promises to the government for fees it would pay over the years were abandoned.

The government announced that FirstGroup (owner of Greyhound and BoltBus) would in December take over operating services from Virgin on the West Coast Main Line, Britain’s most popular, and profitable, railway line. But Virgin, which was hoping to renew its contract, warned that the deal was a “recipe for bankruptcy” because of FirstGroup’s inflated estimates of future ridership, which were used to determine how much FirstGroup would pay National Rail over the next few years to use the tracks. Last week, the government admitted it had made a serious mistake, revoked FirstGroup’s winning bid, and refunded all four bidders for their work upwards of £5 million. This was an unnecessary loss of taxpayer funds and will force the contract to be re-bid in the next few months.

These circumstances are revealing of the problems with separating ownership of tracks and operations. Track owners want to maximize the amount of money they can charge companies to run their services there, and so in places with profitable operations, they will accept the highest — and potentially riskiest — bids. Meanwhile, operators have an incentive to maximize profits — either in terms of inflating estimates of future performance in order to win a bid (in the case of the current U.K. system) or skimping on services or maintenance — in order to pay the operation charges to the track owners.

It is hard to see how mainland Europe’s approach, the first alternative, would be much better. The track owners are setting standard prices for track use by each individual train, which will likely encourage intense competition on the heaviest-traveled routes but a lack of interest in operating services on lines with less passenger travel. This will reduce revenues per train on the popular corridors, making any kind of cross-subsidies currently instituted by national operators less likely and probably requiring increasing government aid for the continuation of services to less popular areas.

Thus the explanation for Stephen Smith’s call for privatizing in the mode of Japan’s railway; by integrating track ownership and passenger services, conflicting incentives can can be negotiated, rather than fought out. Ed Glaeser, the Harvard economist, wrote Sunday that potentially profitable lines in the densest sections of the U.S. — including the Northeast Corridor and parts of California — should “follow Conrail into unsubsidized privatization.” This implies, like in Japan, giving one company rights over both tracks and the trains that run on them.

The problem with this approach is clear: It is a recipe for monopoly control of a railroad by a private enterprise. Conservatives berate Amtrak and other government-owned enterprises for being “Soviet“-like (in the words of John Mica), but private monopoly control of rail services is worse — and market conservatives should agree on that fact. After all, monopolized services can inflate prices, provide poor service, and in general be unresponsive to customer concerns. Unlike Amtrak, which must respond to political demands in our democracy, private monopolies must respond only to their profit-seeking shareholders, who clearly have different demands than the public as a whole.

Some might argue that rail services in the U.S., even if controlled by one company, cannot constitute a “monopoly” since they are competing with air and road services, which are owned and operated by other entities. Yet if it is in the public interest to encourage fast, relatively inexpensive services on intercity rail lines (that is what we want, right?), we are effectively arguing on behalf of massively increasing rail share on specific intercity travel markets — and significant government subsidies dedicated to investments in new tracks would back this approach. Once we have done that, we do not want people to move back to cars or airplanes, and we will have made those alternatives quite unappealing — thus benefiting the monopoly. Japanese rail operators do not have to compete with air or auto travel for most of their services because of the extraordinary advantages of rail along their routes, allowing them to act as transportation monopolies.

The presumptive development of vertically integrated private enterprises for the provision of new rail services in Florida between Miami and Orlando and in the Southwest between Victorville and Las Vegas has led to excited speculation that there is a potential for future American private investment in intercity rail. These investments would produce rail monopolies, but their overall market shares — at least at first — would make them small enough to avoid the prospect of developing into transportation monopolies. So they offer considerable benefits and likely should be supported by the public sector if possible — especially since neither Florida nor Nevada seem likely to be promoting state-sponsored public intercity rail projects in the short term.

Yet as a national approach the vertical integration of railroads into privately controlled monopolies is problematic policy that could encourage higher ticket prices and little concern for the overall public interest. There is an argument to be made that effective regulation of such services, much as is done with private energy utilities in most of the country, could limit the profits and destructive effects of such monopolies. But utility regulation has hardly prevented price gouging among energy companies. Can we trust that private railroads in the U.S. would work as well as those in Japan?

Moreover, private companies will only be interested in operating lines that are profitable. The reason Amtrak continues to serve places like Montana and Kansas is not that the agency has fooled itself into thinking that these can be profitable routes, but rather that its political support from the federal government has been premised upon the continued operation of trains to places that do not “deserve” it. But there are few transportation options available for these areas, and there is a political concern for continuing operations there. Would it make more sense to continue having Amtrak operate those routes, and, as usual, privatize the profits and socialize the losses? Or should such routes be contracted out to the lowest bidder, knowing that such contracts may fall apart, as they have in the U.K.?

If there are significant reasons to be concerned about private monopolies on rail services and there is a consensus that separate control of infrastructure and operations causes more problems than desired, what, then, is the option? Amtrak’s role as a public enterprise with seemingly very little interest in the public itself does not set a particularly helpful precedent. But if a monopoly over all rail services in a specific area is the only reasonable option, is the public sector the more appropriate place to turn than profit-motivated private groups? There are no easy answers.

* The “private” nature of these companies is worth questioning. Of six railway companies, three (Hokkaido, Shikoku, and Kyushu) are fully owned by the government entity Japan Railway Construction, Transport and Technology Agency. In other words, they are public enterprises, like… Amtrak.

** It should be noted that in some countries, like Germany and Spain, the historic nationalized rail company (DB and Renfe, respectively) has a separate division that controls the tracks, on which competing operators can buy running rights.

*** One of the odd consequences of Britain’s privatization schemes (which extend beyond rail and into energy and other services) is a significant expansion of government-owned corporations from other countries entering the U.K. market. Certain bus services in London are provided by Paris Métro operator RATP; certain intercity rail lines are operated by Germany national railroad DB’s subsidiary Arriva. Meanwhile, the U.K. itself has no such state-owned enterprises that have the capability of competing for services in other countries.