Monday, December 1, 2008

"All Of A Sudden We're Popular"

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

Last week the Fresno Bee had a good overview of the funding issues our high speed rail project still faces. The article includes some good details that we haven't discussed before:

Three days after the election, the authority published an update of its 8-year-old business plan. It estimated construction costs at $23.5 billion in 2008 dollars, plus $3.2 billion to $4 billion for trains and about $6.1 billion for design work, rights of way and other expenses.

Besides the $9 billion authorized by Proposition 1A and the hoped-for $12 billion to $16 billion in federal funds, the authority also is counting on $6.5 billion to $7.5 billion from pension funds and other investors, plus $2 billion to $3 billion from local governments along the route.

If it all comes together, the plan says, the authority will be ready for final design and construction in 2012, a little more than three years from now. But that's a big if.

This is the first mention of pension funds, but they are an intriguing and I would argue sensible source of money for the project. The stock market crash has hurt some of the larger public funds, like Cal-PERS, and it makes sense that they'd want a more stable long-term investment. It's worth keeping in mind that it's not anticipated that pension funds would make up even half of the $7 billion or so from private investors, and that the California High Speed Rail Authority has letters of interest from over 40 large investors.

As the US looks to recover from what is already the longest recession in over 25 years sustainable infrastructure projects will be a major part of future growth. It makes sense to use HSR as a 21st century version of the Depression-era projects that did so much to put California back to work. Private investors will find security in our project that has been lacking elsewhere in the economic landscape.

And of course economic stimulus is a major reason why we can expect to see some considerable coin coming out of Congress for HSR in the next year. As the Fresno Bee points out, though, we're not the only ones in line:

An Amtrak reauthorization bill that President George W. Bush signed in October puts the state in competition with 10 other potential high-speed rail corridors for a pool of money currently limited to a modest $1.5 billion. The Washington-to-Boston corridor -- home to Amtrak's Acela, a sort-of-high-speed train that tops out at 150 mph -- is at the front of the line.

California is fighting for second place with nine others, including virtually all of the nation's population centers from Florida to the Pacific Northwest.

That $1.5 billion is the first drips from the faucet - John Kerry is proposing a much larger package for the 2009 session. Still, the Northeast Corridor has powerful allies. There are at least sixteen Senators representing the eight NEC states from Massachusetts to Maryland (damn seventeenth century borders!) as compared to just two from California.

However, California is in a VERY strong political position with San Francisco's Nancy Pelosi as Speaker of the House and Dianne Feinstein and Barbara Boxer wielding significant power in the US Senate. Jim Costa, a member of the House of Representatives from Fresno and the author of the 1998 bill creating the High Speed Rail Authority while a state legislator, is confident that we'll get federal money:

Rep. Jim Costa, D-Fresno, a current San Joaquin Valley congressman and longtime high-speed rail advocate, predicts that any federal help will come in several chunks.

Besides the Amtrak bill, Costa said, funds could come from other legislation on surface transportation and climate change as well as an economic stimulus bill intended to relieve the recent credit crunch and recession.

Lynn Schenk, a CHSRA board member, was also quoted as saying that our possession of state matching funds will put us in a very good position when lobbying Congress - outside the NEC no other state has matching funds for an HSR project (though some states, like Washington, have funds to upgrade existing tracks to reach higher speeds).

The upshot is that California has an opportunity to get HSR funded in 2009. I would prefer that the federal funds come all at once, but Costa is probably right that chunks are more likely. I just hope that doesn't mean we get hung out to dry if control of the federal government shifts in 2012 or later. My own view is that a maximalist strategy is best - we should lobby Congress to contribute the entire federal match at once, and be willing to accept a smaller amount if necessary.

The article also discussed where the first section of HSR track will go:

At some future date, [the Authority] will begin construction on the first section of track, extending about 160 miles from Merced to Bakersfield.

The longest, flattest stretch of the proposed route, that segment is intended for use in testing trains at their maximum speed so that their use can be approved by federal regulators.

Along with expected upgrades to the urban Bay Area and SoCal portions of the route, this suggests that the final pieces are also going to be the most difficult - Pacheco Pass and the Tehachapis. The tunneling work alone will take years, but once completed it should mean that the entire first phase will be open. 2018 is still the ballpark date for SF-LA-Anaheim, and as Kopp pointed out in the article, the sooner we get moving the better.

We won the battle in California for high speed rail this year. Now we must prepare to do battle in the Congress. Our high speed rail project is exactly what this country needs in a recession like this. Let's make sure California gets the federal support we deserve.


Rob Dawg said...

What are you smoking?
This is the first mention of pension funds, but they are an intriguing and I would argue sensible source of money for the project.

Pension funds are specifically PROHIBITED from donating money to causes. It is called fiduciary responsibility. Remember, there will be no "profits" to pay back the capital costs. Any operating surpluses go to system expansion. Remember? There's a word for promising the profits to more than one investor. Remember?

Brandon in California said...

If Calpers participated in the HSR project, it would be an investment, not a donation.

Anonymous said...

Indeed. Capital is capital. Certainly what is being proposed is that pension funds lend money to the project. Interest payments do not depend on the presence of profits.

Rafael said...

@ Rob Dawg -

actually, "operating surpluses" refers to what's left over after debt has been serviced. Any private investment - especially from a pension fund like CalPERS - would almost certainly come in the form of a bond with high seniority but reasonable interest rate.

No-one is expecting them to donate to a cause, though there will be negotiation on just how much ridership risk private investors would have to shoulder. Specifically, if HSR against all expectations fails to achieve fare box returns in excess of 100%, would the debt to private investors still have to be serviced in full, in part or not at all?

Loren said...

Finally saw something on what they expect to build first. This suggests an interim service possibility:

A sort of Super San Joaquin between Merced and Bakersfield.

It would be timed to connect to some of the San Joaquins at Merced; some of the SJ's may even be cut short at Merced. The SJ's take about 3 hours to make the trip; the CAHSR train will take about 52 minutes, over 3 times faster.

Like the San Joaquins, it would connect to a bus from Bakersfield to LA, and perhaps one from Merced to San Jose. Both buses would take about 2 1/2 hours, and adding 1 hour for SF-SJ, the total SF-LA trip time would be about 5 hours.

And as they complete the tunnels, they can open further segments, reducing the trip time step by step.

Brandon in California said...

A return to CALPERS, assuming they were to participate, depends on the arrangement.

We're assuming if they were to participate, or for any private sector entity, that a return would depend on operating surpluses. However, agrements could be pursued that would provide a guarantee with more risk on the CHSRA or the State. Of course, that would likely cap a return on the investment.

Another arrangement could be CALPERS purchasing some portion of the infrastructure and leasing it back to CHSRA. I think such arrangements are termed "lease buy-backs'. But, I am unsure those are permissible any longer. A similar arrangment by WMATA has turned and bitign them in the butt. The purchaser, Citi Group, now wants their money back and WMATA does not have it.

Anonymous said...

Question for everyone. The article says the section of track from Merced to Bakersfield will be built first. I am assuming this means it will be straight through Fresno. How will this "test track" fit into the plan for the 100' corridor in downtown Fresno and into the plans for a proper Fresno station?

Will they have to rip up any construction done in Fresno and start over? Is there even room for bypass tracks in the Fresno downtown corridor?

Rafael said...

Regarding the NEC: I have no problem with putting federal taxpayer dollars into upgrading that to higher speeds. It's the corridor with the highest population and proven ridership on Acela. California wants the rest of the country to chip in on its system, so both HSR corridors should to be developed in parallel, albeit in different ways. Indeed, the two represent opposite ends of the high speed rail continuum.

The California system must cover a lot of flat ground very quickly in the (relatively) sparsely populated Central Valley to achieve competitive line haul times against flights between SF and LA. Therefore, brand-new dedicated track supporting top speeds of 220mph is appropriate.

In the north east, specifically between New York and Washington DC, there is essentially no opportunity for a straight-shot alignment on dry land any longer. This is especially true if the downtown transit hubs of major population centers in-between are to be served as well, as they surely should be. In particular, Joe Biden will almost certainly insist on a stop in Wilmington, Delaware.

My recommendation would be to add some bypass sections to the NEC, include some airports in the route, at least as optional detours. Those already served well by subway, light rail or commuter rail should be bypassed. See this map for details.

In addition, the NEC should be converted into a rapid rail corridor, i.e. the infrastructure and rolling stock of all operators should be upgraded such that FRA-compliant freight trains and lightweight UIC-compliant passenger/cargo trains can share track with no increase in train-on-train collision risk. Freight operators will likely need some public grants to play ball, but that is still a lot cheaper than creating a brand-new alignment, e.g. in the I-95 median.

Electrification should be 25kV AC single phase @ 60Hz throughout. The NEC currently features three incompatible systems on various sections. This can be overcome with multi-modal power converters in the trains, but those are not off-the-shelf. It's better to bite the bullet and implement a sensible national standard for all candidate HSR/rapid rail corridors, because some of them will be linked up in the future.

Passenger trains should be as lightweight as possible to maximize acceleration/deceleration performance. That way, slowing down for tight curves has less of an impact on line haul time. It also makes it easier to add stops to the route if there is political pressure to do so. The Acela Express has to make 13, my map shows just five.

In addition, rapid rail passenger trains for the NEC should use either passive or active tilt systems to maximize speed through corners. Tilting is needed primarily to ensure passenger comfort and sense of safety at high speeds through tight curves - trains become vomit comets long before they actually fly off the rails.

Talgo is the only manufacturer with a proven passive system, i.e. the suspension mount points lie above the center of gravity. Their BT model, popular in Europe, is rated at 220km/h in commercial service (where track conditions permit).

Active tilt systems, offered by a number of manufacturers, permit greater lean angles. One of the most successful is the X2000, in service in Sweden.

The Acela Express also uses active tilt, but the requirement to meet asinine FRA passive crash safety regulations have resulted in sky-high operations and maintenance costs. Hopefully the incoming Obama administration will appoint an FRA chief who understands that the organization can no longer act as a champion of freight operators to the detriment of affordable passenger service.

Unlike typical sensor-based acctive tilt systems, the JR 281 - an excellent DMU design used for regional express service in Japan - uses an accurate track geometry database to anticipate the maximum possible speed and lean angle for upcoming track sections. For historical reasons, JR is required to use rolling stock with just 600m brake distance on legacy track, especially narrow gauge alignments. This limits the JR 281 to a top speed of just 130 km/h, but line haul times are still very respectable because these trains can maintain high speeds even in tight curves and in adverse winter weather.

The JR 281 is a good example of why HSR/rapid rail advocates should focus on actual line haul times rather than the rated top speed of the equipment - a number that lazy journalists and John Q. Public tend to zero in on and grossly misinterpret.

To exploit full tilt capability in the NEC, there will be some curves in which the track will have to be moved laterally and/or the superelevation of the outside track increased. Proper design, as-built documentation and maintenance of track geometry is critical to safe operations. Fortunately, the NEC is actually owned by Amtrak, which makes it that much easier to spend public money on improvements to the infrastructure.

Preventing damage to these upgraded tracks implies reducing the permissible axle load of freight trains, i.e. operators may have to switch to UIC-compliant locomotives and a single container per flatbed car. Significant engineering effort will have to go into finding the best possible compromise between high speed passenger and heavy freight train operations.

Other HSR candidates, e.g. Florida, Texas, Midwest, Mid-Atlantic and Pacific Northwest may or may not follow in the footsteps of these two pioneer systems. Each will leverage the experience gained with both very high speed and tilt-based approaches. One size does not fit all.

Rafael said...

@ Joseph -

you touch on a thorny problem. Fresno wants to shoehorn all heavy rail traffic into a single corridor only wide enough for four tracks. HSR needs two dedicated tracks, both for speed and to comply with FRA rules prohibiting mixed traffic. That means veryone else - UPRR, BNSF and Amtrak - would have to share the other two, a solution akin the to Alameda corridor in Los Angeles.

Things get particularly thorny near the downtown station, where express bypass tracks are needed for both freight and HSR - a total of eight or more tracks, in addition to platforms.

One option would be to stack tracks vertically, such that HSR tracks run on an aerial structure immediately above the freight tracks within city limits. This would permit 8 tracks inside the 100' corridor. This is important because HSR trains need to vacate the express tracks well before reaching a station they will stop at. At the station, there may be a need for more than one platform in each direction, which means acquiring land for additional tracks on both levels.

Passengers transferring between Amtrak and HSR would need to use a concourse level located above the HSR tracks, straddling the ROW. Pedestrian access to the station would be possible from either side.

The other option is to build bypass tracks for freight around the city's western edge to eliminate noise. That would imply the HSR tracks have to run east of the freight tracks, flying over them in the connectors to the Bay Area spur near Chowchilla. That should be doable.

Perhaps the biggest issue in all this is that UPRR and BNSF are competitors. They generally hate the idea of giving up dispatch control to an impartial entity. Also, Amtrak will want to change its San Joaquin route from to the UPRR alignment so the HSR stations in Modesto and Merced become intermodal. The latter is particularly important because it will take many years before ground for the HSR spur to Sacramento is broken.

An obvious option would be to extend a shared freight corridor through Fresno north as far as south Stockton using a short new connector section to cut over from the UPRR to the BNSF alignment. Between Escalon and Stockton airport, such a connector already exists. The problem is that the BNSF alignment is currently single track, so using it as a joint freight corridor would mean laying at least one more track. A somewhat funky but cheaper alternative would be to keep using the UPRR ROW for southbound freight trains and redirect only the northbound ones over to the BNSF track. This would also ease conflicts with Amtrak, as it would get a track all to itself. By adding another track only at stations, this would permit punctual if not particularly fast Amtrak service in both directions (cp. SMART).

In addition to moving the Amtrak route over, smaller towns will also want stops. This is consistent with the notion that the role of the San Joaquin will change from intercity to feeder service once HSR service is available. It would also make the single-track segments shorter, so Amtrak could run its San Joaquins more frequently and with shorter dwell times.

To keep its fuel costs from exploding, Amtrak will have to switch to lighter DMU equipment - FRA-compliant if need be. It should choose the most powerful drivetrain configuration available to minimize the distance required to get back up to speed.

In particular, it should look into hybrid DMU platforms which enable both brake energy recuperation and extra power for accelerating away from stations. Energy storage should based on either rapid charge lead-acid batteries with sufficiently long cycle life, ultracapacitor modules or superflywheels. All of these technologies are fairly expensive and new, though.

Andrew said...


The JR 281 is an interesting-looking train. It would seem that it's only used way up north in Hokkaido. Around where I live in Tokushima, N2000 Series tilting DMU's are used for limited express service.

Brandon in California said...

This AM Obama and Biden were at the national governor’s association meeting. They held an interview with the governors to tell them that the new administration will work with them and listen to them to find and locate the best methods to help their states in getting out of economic downturn. It was more of a listening exercise on the part of Obama and Biden, but I took note of the pre-launch comments by Biden and Obama; particularly Biden.

I believe Biden is tasked with discussing more of the fine grain matter whereas Obama will speak more generally; however, they are both said to be of the same mind on the issues.

And to that, each spoke to infrastructure investment. Obama cited that we (the US government) have spent less than 1% of our _____ (annual GDP?) toward infrastructure.

This indicates that his administration will focus efforts on infrastructure.

Roadways, bridges and such were cited at one point.

Biden spoke to rail in general terms. He cited that he has a rail bias, either that that is true or that others label him as such, I do not know. He also cited that 3rd world countries have invested in new ways to move people, maglev was cited, and they are more efficient. Either Biden or Obama said we are behind those countries and there is no reason for that and we must find ways of doing things better.

That is generally what I picked up… I am going from memory. Fortunately I recorded it and can listen again later.

Anonymous said...

Acela does tilt and there are off-the-shelf multi-voltage HSR, including Thalys, ICE, TGV and the Eurostar.

Anonymous said...

@ Rafael

Wow, I new their was some issues with that alignment, but nothing so severe.

While an overhead passage for the express trains seems like a good idea,wouldn't it have to be above the ground level streets and roads?

Could the system create a double-decker trench? Fresno dosent have so many seismic issues and that seems more viable and acceptable.

You could have the lower layer of the trench with 2 freight lines + 2 HSR Express Lines, with the 4 passenger lines directly above.

Unless I'm missing something important...

Rob Dawg said...

Robert Cruickshank said...
Donation my ass. They would rightly expect a return on their investment.

Exactly. Just how does this work? I'll tell you. They buy a portion of the $9b bond offering. They don't "invest" in some new form of capital debt because there is no new capital debt authorized. You are double counting any potential CalPERS participation as if it were in addition to the initial $9b.

We have the first $9b and it is debt at a rate to be determined. CalPERS is still a highly competent investor and would be a very likely investor in those bonds. But because they are competent investors they aren't going to bring any other form of capital investment to the table.

Anonymous said...

Rob Dawg - No, the quote in the post makes very clear that the pension funds would be purchasing additional debt in the $6-7 billion private financing portion, not in the $9 billion GO bond portion (though they might or might not purchase some of the GO bonds as well...whether they do or not is of little relevance). Whether pension funds will actually participate is an open question, but the quote itself is clear.

Anonymous said...

I expect the Tehachapi connection to be started ASAP after the Central Valley track goes in, or possibly before.

They've already progressed to project-level planning on it, refining the route down to a choice between hundred-yard-wide corridors, and done a *lot* of the EIS work.

The Central Valley - LA travel demand is also huge and largely unserved; it's a good place to try to get service as early as possible.

In contrast the Merced-Bakersfield section doesn't seem nearly as far progressed with the environmental documentation and detailed planning. I suppose they figure it's going to take less time to construct so they can afford to start the EIS stuff later?

Anonymous said...

Rafael - I too think that major investment in the NEC is well-justified. It could be called the NECIP II (Northeast Corridor Improvement Project II). But as someone who is familiar with the corridor, I find your alternative ROWs very odd. You're bypassing what is basically the fastest section on the NEC, from Frankford Junction to Metropark, and then reentering the corridor prior to bypassing the horrid reverse curve at Elizabeth (SR 55 mph, even for tilt trains).

This map overlays what I'd do for the Pennsy (NEC from WAS to NYP). This plan would result in 100+ mph running everywhere except WAS to Ivy City (Washington DC), Arsenal to Zoo Interlocking (Philly), and the approach to NYP. That is okay since WAS, PHL, and NYP would be the three stops that every train makes.

Running times for a WAS-PHL-NYP express should easily fall below 2'10". The expensive parts are new tunnels under Baltimore and under the Hudson. The latter isn't really a project cost because it's going to get built sooner or later anyway, but the former could easily run $2-3 billion. Everything else would probably add up to under $3 billion total.

The New Haven/Shore Line is tougher (i.e. more expensive). I haven't even tried that one.

The proposed changes, going from east to west:

Diverge from corridor at West Baltimore. Use never-completed Hwy-40 median to approach downtown. 3 mile tunnel under Baltimore from MLK Blvd to I-895. This is expensive but there is no cheap way to bypass the super-slow Baltimore tunnels at high speed. The existing tunnels could be refitted to provide additional capacity for MARC and freight. Baltimore Penn might be downsized.

Retrofit Susquehanna River Bridge from 90 mph to 110 mph.

Retrofit bridges from Baldwin to Marcus Hook (Philly suburbs) from 90 mph to 100 mph.

Realign curve at 2nd St (North Philly) for 100 mph (currently 65 mph; real estate purchase necessary - good news is that North Philly is depressed, i.e. land is cheap).

Realign curve at Frankford Jct for 100 mph (currently 50 mph; large real estate purchases necessary - fortunately it's mostly abandoned warehouses).

Realign reverse curve at Elizabeth for 100 mph (currently 55 mph; some real estate purchase necessary - mostly parking lots).

High speed (90 mph) run-through tracks at or around Newark Penn Station (this will not be cheap).

Additional 90+ mph double tracks from Newark Penn Station to east of Secaucus Jct (they need the capacity anyway).

2nd Hudson River tunnel (they're going to build this sooner or later anyway).

Anonymous said...

looking at all the info on the CHSRA libary every article seems to point to a 2012 construction start up and LA-SF to start 2020.
One of the charts has the Pacheco pass portion underconstruction thru
2019. where is 2018 coming from? Unless its a small segment

Anonymous said...

Staging Tehachapi first would permit trains to run from LUS onto the existing SJ route. Even this should attract ridership (i.e. it would replace the AmBus LA-Bakersfield gap). Would there be any time gained in LA-SF using Amtrak California stock?

Rob Dawg said...

mike said...
No, the quote in the post makes very clear that the pension funds would be purchasing additional debt in the $6-7 billion private financing portion, not in the $9 billion GO bond portion.

That's so cool. When do we vote on this new debt? Ans: we don't because this is fantasy. Private financing does not involve repaying capital investments. Read the business plan. The CAHSRA plan involves UPRR and such giving billions in exchange for... well I don't know what they can expect only that they are expected to contribute. There's no paying back part.

Rafael said...

@ Joseph =

a covered trench solution would mean disrupting freight service during construction, plus it would put the lighter HSR trains under the heavy freight trains. It takes more rebar to support 15,000 tons than it does to support 400. In addition, there may be sewer pipes, water pipes, who knows what under those UPRR tracks and certainly left and right of them. Plus, digging is expensive.

Don't go underground if you don't absolutely, positively have to.

@ mike -

I stand corrected, you clearly know that area much better than I do. My point was simply that a California-style system is infeasible in the NEC, so the focus should be on getting the best available lightweight tilt train technology working optimally.

@ yeson1A -

CHSRA had previously indicated 2018-2020 as the timeframe for start of operations. That could mean partial service by 2018 and the first trains running from SF all the way to Anaheim in 2020. If the history of such projects is any guide, it's entirely possible that will still be pushed back some.

In particular, tunneling through complex geology near seismic faults is like a box of chocolates. You never know what you're going to find. Some types of rock will creep high under pressure, a little like toothpaste but much more slowly.

Other strata are like sponges, full of water. In Spain, construction of the Cordoba-Malaga AVE line cost the Valle de Abdelajis region its precious aquifers. What used to be an oasis is now turning into a desert, so CHSRA had better not repeat that particular mistake, especially at the eastern end of the Pacheco Pass route.

Others strata still are full of natural gas - that's what they ran into back when they built the SP railroad tunnels through the Santa Cruz mountains. 32 people died in the explosion that followed. Again, do not repeat...