Thursday, January 22, 2009

California as HSR Model?

NOTE: We've moved! Visit us at the California High Speed Rail Blog.

While we are debating the role of passenger rail in the economic stimulus, our neighbors to the north (Canada, not Oregon) are looking to us in California for direction on their high speed rail concepts. An article in Monday's Globe and Mail explores what we have to offer, and concludes the key point is private partnerships:

The success of Proposition 1a should come as welcome news to Canadian high-speed rail advocates, who have long dreamed of such service between Windsor and Quebec City, and connecting Calgary, Red Deer and Edmonton...

The implications for Canada are clear: If we want to drag Canadian passenger rail service out of the 1960s, Ottawa must include private partners, a development that could mean selling the moribund Via Rail to a transportation consortium that understands the technology, the market opportunities and is willing to invest.

I am not convinced VIA Rail's problems stem from any innate "moribund" problems but from a persistent lack of support from both Liberal and Conservative governments, but the author seems to believe that the great lesson of California is that private partnerships are necessary to building HSR.

I don't think that's entirely accurate. Private investment has been seen as a way to round out the final financing numbers and in particular a way to earn the backing of Governor Arnold Schwarzenegger. The key to California HSR's success is public support - from Proposition 1A to the all-important federal contribution. I have always argued private investors will be interested in participating, but that their role should be limited. John Lorinc, author of the Globe and Mail article, doesn't agree and seems to bring an agenda primarily driven by free market ideology and not evidence:

What's clear is that pure private plays don't work, while publicly owned passenger service tends to suffer from a lack of investment. (An exception is Spain, where the state has moved aggressively to build high-speed rail since 1994.)

Earlier attempts to privately develop such networks in Texas and Florida failed because of inadequate financing.

Publicly owned passenger service has not suffered from a lack of investment in China, France, Germany, or Italy (I'm sure there are other examples out there; those are just the first four that come to mind). And of course HSR in Texas and Florida wasn't failed by financing problems but was killed by the Bush brothers.

Not surprisingly, it is Alberta where talk of privately operated HSR has been centered:

In Alberta, meanwhile, a consortium led by retired banker Bill Cruickshank [no relation] has lobbied Alberta to consider a privately operated high-speed rail link between Edmonton and Calgary, to be financed through a proposed partnership between the consortium and public backers. Alberta Premier Ed Stelmach made positive noises about the plan, but has refused to release the results of a market assessment conducted in 2007. As Mr. Cruickshank notes, "If the government said it wouldn't work, I'm sure they would have told us."

High-speed rail, as Mr. Cruickshank argues, is tailor-made for public-private partnerships. Such train networks are extremely capital intensive and therefore depend on public financing to get going. But as a transportation service that competes vigorously with airlines, private operators will bring the necessary marketing savvy, as well as the ability of amortizing capital costs over a long period.

I can't disagree that HSR and PPP are a good match - given the logic of PPP. Whether that logic is the best possible method of building and operating HSR is another matter entirely. I am pleased that Bill Cruickshank understands the transportation and environmental benefits of HSR and has been a strong advocate for passenger trains in Alberta. And, given the political realities in that province, PPP is probably the only way an Alberta HSR will ever become real.

But that would be a conclusion Alberta draws on its own. That is not the California lesson. No, what California shows is that if you give voters the choice, they WILL commit public funds to high speed rail, and that private investment may have a place but is neither necessary nor sufficient to HSR success. HSR deniers spilled a lot of ink arguing that Prop 1A was a stealth tax increase on Californians but voters approved it anyway - and approved outright tax increases for trains in San José, Marin-Sonoma, and Los Angeles. The Canadian Liberal Party's Green Shift may have been ill-conceived and poorly sold by the inept Stéphane Dion, but given the choice, it is likely that Canadians too would vote to spend public money on high speed rail, even if it cost them more in taxes.

What California shows is that the key is generating public support for high speed rail as a concept. The specific mechanism of funding, from the role of private financing to the generation of public monies, are important but not determinative. Canadian HSR supporters, whether they are in Alberta or the Windsor-Québec City corridor, should focus their efforts on making the case for HSR to the public and in the halls of Parliament and provincial assemblies. Public funding can either bring private supplements or, if the groundswell is strong enough, be itself sufficient for HSR projects.

Now to convince DC Democrats of that fact...


Alon Levy said...

Speaking of Canada, I'm interested in the feasibility of a New York-Toronto line.

Is there any easy way to provide international connections over rail? Right now, trains have to stop at the US-Canada border for two hours for customs and immigration checks. The Cascades service has the checks done airport-style in Vancouver, after which passengers board the train from a secured platform, but that only works because that the train makes just one stop in Canada, so that there's no intranational service in Canada that this arrangement impacts. Without some way of expediting border controls, trains will never be competitive between the US and Canada, where airline border controls take twenty minutes.

Rafael said...

@ Alon Levy -

before the Schengen accord eliminated internal border controls in much of continental Europe in favor of distributed immigration and customs enforcement, it was common for immigration officials to board trains and inspect passports while they were moving through the areas close to the borders.

The one exception I personally experienced was between Greece and Turkey. Officials collected passports and thoroughly searched the one rail car that was permitted to cross the border for contraband. Everyone else had to disembark and wait for carriages from the other country's railway to arrive. After several hours' delay, the immigration officials walked the train with stacks of passports and returned them to their owners.

The US, like the UK, also maintains an immigration system based primarily on robust inspections at its borders. The events of 9/11/2001 only made matters worse because immigration officers now need not just bits of paper but also fancy fingerprint equipment and databases to process their customers.

One possible approach would be to assume that all passengers boarding a train that is headed across the border will actually cross it (otherwise they should board a different train). That means all of them have to pass through immigration and customs somehow, possibly including metal detectors X-ray scanning of their belongings.

There are several options for processing: prior to boarding, en route to the border or, at the border. The option that minimizes delays for passengers is processing en route, but it is also the most difficult to implement. The UK therefore relies on checks prior to boarding for departing passengers and after alighting for arrivals.

That means each and every station served by Eurostar trains to and from France has to be treated as a border crossing, conceptually equivalent to an airport. Trains must use dedicated platforms that force pedestrian flow through the check points. There is currently no true HSR service within the UK but there will be once HS2 is constructed. Those domestic trains will use regular platforms to bypass customs and immigration.

My guess is that cross-border HSR service between the US and Canada (Niagara Falls, Detroit/Windsor and Blaine/White Rock) or Mexico (Otay Mesa/Tijuana) would all emulate the UK model.

Alon Levy said...

I think I understand. So, for the NY-TO corridor, there would be three types of service - from NYC to Niagara Falls, NY; from TO to Niagara Falls, ON; and from NYC to TO with all US stations used only for boarding and all Canadian stations used only for alighting?

Anonymous said...

My understanding of the idiocy of the Eurostar's security procedures is that they're related to safety concerns of the Eurotunnel, not border controls. DB has expressed its intention to bid to operate to London when open access comes to the EU rail system in 2010. Others, including Air France, are also eying operating trains from further afield to St Pancras.

If trains now are coming from as far as Amsterdam and Cologne, I doubt stations will be adapted to provide the secure areas of the current Eurostar. As the trans-alpine tunnels open, there will be more very long tunnels with a lot more trains running through them. The paranoia that drove the Eurostar's security procedures will be rationalized into more open service.

As stated before, if we need to look at border controls between the US and Canada, inspectors on trains is better than a two-hour wait. Two hours? That's a market killer.

Rafael said...

@ Alon Levy -

since Niagara Falls is a destination in its own right on either side, running some trains only as far as the border might make sense.

For a train from NY to Toronto, anyone boarding on the US side would be expected to cross the border. They could theoretically alight before reaching it, but would do so on the same platform that others use to board - i.e. one that requires passing a customs and immigration post to leave the building. To avoid that hassle, they should just take a domestic train instead.

Once the train crosses the Canadian border, it would be considered a domestic train there. Canadian residents traveling to Toronto would be permitted to board at any of the remaining stops.

Note that SNCF operates semi-express trains within France that run as local trains for one-half of the journey and non-stop for the other. Passengers traveling between secondary destinations transfer at the mid-point. This model could be applied to US-Canada cross-border trips as well:

a) stop at each station between NY and the Canadian side of Niagara Falls, then continue non-stop to Toronto. Passengers to secondary destinations in Canada would have to transfer.

b) proceed directly from NY to the US side of Niagara Falls, then stop at each station to the final destination. Passengers from secondary origins in the US would have to transfer.

Roughly equal numbers of types a and b would be operated, with equivalents on the return leg.

At roughly 530-550 miles (via Albany), there might also be a market for non-stop HSR trips from either NY or Boston to Toronto, especially if reliable broadband internet access is available throughout. Travel time could be as little as 3.5 hours, plus ~20 minutes for Canadian customs & immigration. For comparison, line haul time for flights between NY and Toronto is around 2 hours, but that's without ground transportation, check-in, security check and baggage claim.

Non-stop sleeper car HSR services that do these runs in ~8 hours might be popular as well.

Rafael said...

@ anon @ 11:16am -

UK immigration and customs procedures are defined by the politicians, not any railroad operator.

Eurotunnel also requires security checks for passengers boarding on the continent to travel to the UK, because there have been three fire incidents in the tunnel since it opened, though none involved Eurostar trains. However, SNCF has invested in check-in capacity. Typically, passengers spend less than 15 minutes in a queue.

Within continental Europe, there are no immigration checks at all for travel within the Schengen zone.

Alon Levy said...

There's definitely going to be a market for NY-TO HSR. It's not exactly true that HSR is only competitive at 3 hours or less. As far as I can tell, a better determinant of success is whether station to station time is at most an hour and a half more than for air; the 1.5-hour penalty represents average time to get to and from the airports, check-in, and security. In cities with congested airspaces and inconveniently located airports, such as New York, air gets a further penalty, of about half an hour.