DOT Finance India Metro Rail

Indian Commitment to Heavy Rail Adds Perspective on U.S. Funding System

Indian national government would finance 50% of locally-approved metro rapid transit projects in cities with more than one million inhabitants. The United States, meanwhile, has no set policy on how to finance public transportation.

Jeff at the Overhead Wire pointed to the Indian parliament’s recently approved bill that will provide a 50% commitment from the national government for any metro rail projects that have received a 50% financial guarantee from their respective cities. The U.S. government, on the other hand, has been unable to establish similar uniform standards that define how new transit projects are financed. India could provide a useful example to demonstrate how funding relationships between multiple levels of government can be standardized.

Apart from China, India’s cities collectively have the largest heavy rail transit expansion program in the world. Of the 42 cities in the country with more than one million inhabitants, thirteen are advancing the transportation mode. The Delhi Metro opened in 2002 and has been wildly successful, serving one million riders a day; the system will expand to 200 km of track by 2011. Bangalore, Chennai, and Mumbai have lines currently under construction, while Ahmedabad, Chandigarh, Gurgaon, Hyderabad, Lucknow, Kanpur, Kochi, Kolkata, and Pune are all actively planning the construction of major new lines.

The Delhi Metro was constructed with a 50/50 national-state financing arrangement, with half of funds coming from local authorities. The bill passed by the Indian parliament extends the same promise to any city with more than one million people that is able to assemble a financial package that covers half of the project’s cost. The parliament doesn’t appear to have placed any limits on this obligation, and it could theoretically finance systems in all 42 of the qualified cities. There is no cost-effectiveness requirement for new lines, which means that if cities are able to come up with their half of the funds, any project, no matter how complicated or expensive, will get national government money.

There is no such guarantee in the United States. That said, American cities are working hard to build up their transit systems, which means that there is a need for the federal government to be more clear about how much money it will be willing to contribute to new systems. Today, there is no such clarity:

Recent U.S. Federal Commitment to Heavy Rail Metro New Starts
Project Cost Federal Commitment Ridership/Mile* Cost-Mile/Rider // Cost/Rider *
New York City Second Avenue Subway Phase I
$4,867 m 27% 87,000 $56 k // $24 k
Washington Largo Metrorail Extension
$607 m 60% 6,500 $93 k // $30 k
Washington Dulles Corridor Metrorail $3,142 m 29% 7,300 $430 k // $37 k
Chicago Ravenswood Line Renovation $530 m 46% 7,500 $71 k // $8 k
San Juan Tren Urbano
$1,250 m 25% 10,600 $118 k // $11 k
San Francisco BART Airport Extension
$1,510 m 50% 8,500 $178 k // $20 k
Los Angeles Red Line North Hollywood Branch
$1,311 m 52%
Chicago Douglas Line Renovation
$483 m 66% 4,100 $118 k // $18 k
Atlanta MARTA North Springs Extension
$463 m 66% 14,300 $32 k // $14 k

* These are not “real;” rather, they’re pre-opening predictions calculated by the respective transit agencies and approved by the FTA. If the FTA were basing its financial commitment on ridership or cost effectiveness this number would presumably be relevant.

In the last ten years, the FTA has doled out billions of dollars for new rail lines, with the largest commitments generally going to heavy rail construction or expansion. Like with other New Start projects, from bus rapid transit to light rail lines, however, the U.S. has arbitrarily contributed between 25 and 66% of the cost of projects. Once a project has been approved for national financing by making it through the New Starts process, Washington will pay something, but each project’s sponsor much negotiate its own deal.

The results are inexplicable. There is no clear correlation between federal government responsibility and total cost or ridership per mile, or even cost effectiveness. What appears to be happening is that representatives from states and cities go to Washington and hope to get the best deal, and then the FTA makes a financing decision that has nothing to do with relative merit. In terms of per person benefits, the construction of New York’s Second Avenue Subway may be more important than that of any other transit line in the country. Yet the corridor only has a 27% commitment from the FTA; on the other hand, the short extension of Atlanta’s MARTA finished earlier in this decade got a 2/3 sponsorship. Why? Similarly, based on the numbers above, the San Juan Tren Urbano and the Chicago Douglas Line Renovation would cost the same to build per rider-mile, yet the feds allocated 25% of the price to the first and 66% to the second.

The problem with this system is that it makes it very difficult for cities to accurately predict how much money they’ll have to raise from local sources, and long-term plans are as a result often inaccurate. A system such as India’s, simplistic as it might be, at least would make clear that if Houston wanted to build a $2 billion rail line, it would simply need to raise $1 billion — and then the federal government would fill in the rest.

That, after all, is roughly how the Interstate System was built. Congress authorized about 50,000 miles to be built, and when a state got around to building a section, the Federal Highway Administration would simply distribute 90% of the necessary funds — no matter how complicated or “wasteful” the project’s specifics turned out to be.

India’s example isn’t necessarily ideal: we should expect a measure of cost efficiency on the part of new transit construction, though it is worth pointing out the FTA’s existing guidelines don’t always produce the most effective projects. And the U.S. tax system, which concentrates most discretionary spending at the national level, suggests that a 50% commitment is too small.

But this act of the Indian parliament does suggest that the Congress has an obligation to reform the manner in which transportation dollars are distributed. States and cities should be able to rely on funding in association with the quality of their respective projects. A system in which better, more effective projects get a larger national cost contribution seems ideal. The existing financing conditions should not be a model for the future.

9 replies on “Indian Commitment to Heavy Rail Adds Perspective on U.S. Funding System”

The table you give is confusing. The appropriate comparison to total cost is total ridership; ridership per mile is more relevant to cost per mile. If you compare ridership per mile to total cost, then the MARTA extension looks like a bargain.

At least the Feds are not giving too much money to the two richest counties in the nation (Fairfax and Loudoun) so they can get their Dulles extension boondoggle.

What does ridership/mile mean? The Dulles Metro corridor will be some 10 miles long to begin with. Does that mean 73,000 estimated riders a day? Once it is built out to the airport I think the numbers will surprise. Tyson’s Corner is also a massive job center.

I think a set funding commitment of 50% (or whichever higher value) would be vastly better than what exists now, precisely due to the simplicity and certainty. It would basically eliminate a level of politics, which can only be good for getting things done.

Would specific federal requirements for cost-effectiveness even be of value within such a context? Any city that wants to build a transportation project would know that it will have to pony up 50% of the cost itself, so it has every reason both to minimize costs and to maximize benefits. (The only issue I can see is that richer cities would be able to build larger projects and consequently get more money from the federal government — but I doubt this would be a large factor.)

Yonah, just one more quibble: the metric you added isn’t cost per rider mile but cost-mile per rider. Doubling the length of a line without changing cost or ridership means that ridership per mile halves, which doubles the cost metric rather than halves it.

Alon –

Not a quibble at all, you’re absolutely right.

The advantage of the cost-mile/rider metric is that it is biased towards denser systems, which I think is important when we’re talking about efficient transit. (In other words, a system with the same ridership and cost but a longer distance is considered less valuable than one with a shorter distance.)

That said, I also added cost/riders because that provides a significant other piece of data that’s important for comparison’s sake. Now, if only those high ridership estimates in San Juan and San Francisco were actually backed up by reality…

There’s no doubt that SAS has more ridership potential than an extension in the suburbs, sure. For one, SAS can be extended to connect to more things – e.g. even Phase 1 would be enough to shorten my current commute from 50 minutes to 30, using the transfer at Times Square (not that I’m going to live on the East Side and work on the West Side in 2030 or whenever that line opens, but still). But it’s still by far the most expensive subway anywhere. Even in Tokyo subway construction costs one third as much per km.

The advantage that India has is the near complete absence of any market risk, as a consequence of its immense population, and the current lack of services. As long as the services are designed reasonably there is absolutely no reason for them not to enjoy excellent patronage. This is possibly the reason that the central government did not insist on efficiency measures in the legislation.

Leave a Reply