Commentary

California HSR: Limited Clear

Written by David Peter Alan, Contributing Editor
image description

Exactly three years ago, I pronounced the California High-Speed Rail project dead, and issued a poor prognosis for high-speed rail generally in the United States (Whither (Wither) High-Speed Rail?, posted here in February 2019).

Things looked terrible for the project then, especially since the FRA was about to claw back a $929 million grant because it did not look like the project would be completed by its original projected date in 2022. By that time, also, the projected cost had grown from $32 billion to $70 billion. Today the cost could go as high as $105 billion but, all in all, the proposal appears to be in better condition than it was three years ago. Construction is proceeding in the Central Valley and its proponents, especially the California High-Speed Rail Authority (CHSRA), are hopeful.

Americans have often shown some confusion about what “high-speed rail” is, or isn’t. David Peironnet, who started his career working in passenger service on the Santa Fe in the early 1970s, once told me, “High-speed rail is the schedule that the Santa Fe ran in 1962.” Philadelphians have talked about the “Norristown High-Speed Line” and the “PATCO Speed Line” for decades. On the other side of the spectrum, The Texan reports that local officials have expressed a preference for high-speed “rail” at a velocity of 250 mph between Dallas and DFW Airport over an Elon Musk-style “hyperloop,” an obvious example of overkill.

We have recently covered developments about other high-performance rail projects in the nation: Brightline in Florida, Brightline West in Southern California and onto Las Vegas, and the Texas Central proposal, which is heading straight for a legal red signal. While these other projects do not meet world standards for “high-speed rail” and come from the private sector, the California project is different. It would reach genuine HSR velocities and comes from the public sector. Change is taking place with respect to the other projects, as well as the California project, so it’s time for an update. 

Railway Age has reported about the project from time to time over the past three years, usually when an EIS (Environmental Impact Statement) for a certain segment was released for public review and comment before being adopted, as is required for projects of this sort. They were San Francisco, San José and south (posted Sept. 19, 2019, further reporting posted July 9, 2020); San José to Merced (posted April 23, 2020); Burbank to Los Angeles (posted Sept. 23, 2020); rising costs (posted Sept. 23, 2021); Gov. Gavin Newsom’s decision to concentrate on the segment between Merced and Bakersfield (“California Roars Back”? Nope!, posted May 18, 2021); and two stories posted on Feb. 10, 2022: a report on the release of the project’s new business plan, and the release of the final EIR (Environmental Impact Report)/EIS for the 14 miles between Burbank and Los Angeles. The next day, Editor-in-Chief William C. Vantuono posted a commentary entitled Why is High-Speed Rail Such a Heavy Lift?, which dealt with the national scene and not California specifically.

Against this backdrop, I examined where the California project stands now. In a nutshell, it is better off than it was three years ago, although that might not be saying much. The project is moving forward, and construction is proceeding in the Central Valley. There is funding for now, although sufficient continued funding could become problematic later. It will also be at least 11 years before any trains roll, and more years before the Authority expects to get out of the Central Valley and go south of Bakersfield toward Los Angeles. They have their reasons for that, which I will describe. Whether or not their decision will be remembered as optimal in the long run, despite their reasons, remains to be seen. For now, though, the bottom line is that the project is moving forward, and even if signals are yellow rather than green, one of the goals is to build an operable segment of FRA Class 9 track, with a maximum operating speed of 200 mph. If the plan succeeds in the Central Valley, even if nowhere else, that part of the line will be the first segment of such track in the U.S.

Where the Line Would Go

The 2022 Draft Business Plan was released on Feb. 8 and can be downloaded below. It is a substantial document, 103 pages long. There is a 60-day comment period that ends April 11. 

An important feature of the document is a map of the proposed system. Phase 1 of the system, slated to begin service in 2033 if funds are available (but that is not a foregone conclusion) will be a line from San Francisco to Anaheim through Los Angeles, which would fulfill the project’s core mission. It would start in San Francisco and proceed southward on the Peninsula, on the current Caltrain line to San José (where the full-service part of the line ends, 43 miles), and further to Gilroy, which has only three weekday round trips for peak-hour commuters. The line is being electrified for HSR operation, and for future Caltrain service. At Gilroy, the route would turn left to proceed through Pacheco Pass and then east to the Central Valley segment (88 miles to Carlucci Road, east of Pacheco Pass), with a wye going northbound to Merced or southbound to Bakersfield (33 miles between the two towns and another 28 to the wye on the east-west alignment). Current plans call for construction in the Central Valley to be completed first, as far as Bakersfield (138 miles from Madera). From there, the line would extend through the Tehachapi Pass to Palmdale (79 miles), and near Metrolink’s Antelope Valley Line through Burbank (41 miles), and to Los Angeles Union Station (13 miles). Palmdale is also slated to be a stop for Brightline West trains to Las Vegas, which would create a transfer point between the high-speed line and trains to the gambling mecca. South of the City of Angels, the high-speed line would proceed through Fullerton and terminate at Anaheim, on or near Metrolink’s Orange County Line (31 miles). The entire line would be 494 miles long. The most expensive segments are San José to Carlucci Road ($13.6 billion, including Pacheco Pass) and Bakersfield to Palmdale ($18.4 billion, including Tehachapi Pass).

Phase 2 calls for two additional segments. North of Merced, trains would stop at Modesto and Stockton, terminating at Sacramento. This segment would essentially run parallel to the northern portion of Amtrak’s San Joaquin line, where most trains go from Stockton to Emeryville in the East Bay and a few go to Sacramento. At the south end of the line, Amtrak trains, as well as local trains on Metrolink and Coaster, will continue to go from Los Angeles to San Diego along the Surf Line on the coast. The high-speed line would use another alignment from Los Angeles eastward to San Bernardino, and then southward to Riverside, Murieta, Escondido, and terminating at San Diego. The Authority is focusing on Phase 1, so there are no cost or schedule estimates available at this time. Presumably, there must be sufficient funding in place to complete Phase 1 before estimates for Phase 2 can be determined.

Kyle Simerly, Public Information Officer for the California High-Speed Rail Authority, told me that some of the segments are close to completion: “San Francisco to Merced sections are expected to go to the board for potential approval this year, and the Palmdale to Burbank section next year. Meanwhile, we’ve already cleared many other project sections, totaling nearly 300 miles of the 500-mile Phase 1 system between the San Francisco Bay Area and the Los Angeles Basin and Anaheim, via the Central Valley, including this year the first miles in Los Angeles County since we began.”

The Central Valley

This part of the line deserves some attention, because that is where construction is ongoing. Plans currently call for stops at Merced, Medera, Fresno, a station to be called Kings/Tulare and Bakersfield in the initial operating segment. A station at Madera will be added later, but will still be part of Phase 1. The segment between Madera and Poplar Avenue, 19 miles north of Bakersfield, is under construction, while environmental documents have been completed for the rest of the way to Bakersfield and the segment between Madera and Merced, as well as the east-west alignment between the Central Valley line and the Pacheco Pass.

Madera to Poplar Avenue, north of Bakersfield, is the first construction zone, 119 miles long. The Central Valley segment will eventually extend 171 miles, from Merced to Bakersfield. According to Simerly, it should take about 80 minutes for the ride along that segment. That would average out to 131 mph, including stops and associated accelerations and decelerations.

The planned route is close to that followed by Amtrak’s San Joaquin route, historically on the Santa Fe, now part of BNSF. That route is 168 miles long, and current Amtrak trains take three hours (plus or minus ten minutes) to make the trip, at an average speed of about 60 mph. There are some differences between the routes with respect to communities with stations: there is no station comparable to Kings/Tulare on the current San Joaquin line, but it has additional stops at Hanford, Corcoran and Wasco, all between Bakersfield and Fresno, where the high-speed trains will not stop. While the two lines will be near each other, Simerly explained the difference: “Where possible, to minimize impacts, we hew closely to already established transportation or utility corridors, per statute. But deviations from those corridors are sometimes necessary because roads, ordinary railroads and utility corridors can make sharp turns, where our trains cannot, if we are to maintain our voter mandated speeds.”

The Funding Picture

As often happens with infrastructure mega-projects, costs keep escalating. The California HSR project is no exception. From $32 billion at the start, the 2018 business plan listed the “high” estimate as $98.1 billion (at 36), the 2020 plan at $99.9 billion (at 108) and the newly-released Draft 2022 business plan at $105.1 billion (at 69). All cost figures are in year-of-expenditure (YOE) dollars. More about that later, but it seems reasonable to use the “high” numbers these days, in light of current inflation levels. Of course, the Authority hopes that the cost will be lower than that and has included such figures. The question is obvious: Will the project be built in its entirety, considering the present situation, with skyrocketing costs and the prospect of more inflation ahead? Three years ago, I reported that the expressed intent of the FRA to claw back $929 million in grant funds appeared to be the proverbial straw that broke the back of the proverbial camel’s back, even though the amount at stake represented only 0.88% of the now-projected total.

On June 11, 2021, the FRA reversed itself and reinstated the grant. A press release included a statement by CHSRA CEO Brian P. Kelly on the occasion: “With this settlement, it’s clear we once again have a strong federal partner on this challenging but transformative project. We appreciate FRA’s expression of confidence that we are getting this project on the right track. Let’s continue the work of creating jobs and building the nation’s first truly high-speed rail project right here in California.” 

It is easy to be skeptical about money, but many transit properties are having trouble keeping up the payments for their operations, in light of diminished ridership since the COVID-19 virus struck almost two years ago. This has strained the capital side of budgets, too. While the California HSR project is novel in this country, the sort of infrastructure that it requires is very expensive, regardless of where HSR is built. Rail managers and advocates for riders have dealt extensively with budgets, in an effort to find the money for operations and capital expenses in many places around the nation. Simply stated, it’s not easy.

At an estimated cost of $105 billion for Phase 1 now, I asked how it could be done. Simerly told me: “The project has never been fully funded, and as such we must complete what we have funding for now and continue in this building block approach (as other mega projects are built) as funding becomes available. But we do estimate we actually currently already have all the funding confirmed or authorized to complete Merced to Bakersfield. This section’s cost estimate is inclusive of complete statewide environmental clearance of the entire Phase 1 system from San Francisco to Los Angeles/Anaheim, initial track, systems like signals and electrification, initial stations, initial trainsets, testing and certification of those trainsets, and crucial coordinated preparatory investments in ‘bookend projects’ like Caltrain electrification and the LA Union Station renovation LinkUS project. That section has a total base estimated cost between $22.5 and $24 billion. You can see this on the chart on page 69 of the Draft 2022 Business plan.”

Regarding the commonly accepted total cost figure of $105 billion, Simerly said: “The $105 billion number you focus on is at the very high end of our estimated cost range. For Phase 1, the 500 miles between the San Francisco Bay Area and the Los Angeles Basin and Anaheim, via the Central Valley, the current cost estimate ranges from a low of $72 billion, to a base of between $86 billion to $88 billion, to the high of $105 billion, all in YOE (Year of Expenditure) dollars, which are used in capital cost estimates for public infrastructure projects where construction spans multiple years and are used explicitly to account for the effect of projected inflation on costs over a projected delivery schedule. Cost increases reflect new benefits and mitigations communities will receive where we’ve finalized our environmental documents. This means the project is now better integrated into the communities through which we will travel.” He mentioned a few: noise barriers, safety enhancements, an alignment that will avoid the Caeser Chavez National Monument, mitigation measures to minimize disruption of air-rail connectivity at Burbank Airport, and other community improvements.

Simerly also mentioned the importance of the federal funding component, alluding to a 50/50 match: “Going forward, we’d probably hope for something close to parity with the state’s investments. California is well-positioned for any federal funding approach for passenger rail, and cap-and-trade funding already committed to the California high-speed rail project could be one source for California’s matching share. Given this, there is a strong case that additional future federal participation in the California high-speed rail program is warranted, starting with additional funding for the Silicon Valley to Central Valley line because it would leverage a significant increase in ridership, connectivity among major urban centers, revenues, and the value of private sector concession agreements. This is consistent with historical precedent where the federal government plays an important role in funding large infrastructure projects. Brian Kelly has highlighted half a dozen federally competitive grant programs, recently infused with additional funding through the Bipartisan Infrastructure Act, adding up to $50 billion-plus. The Authority will aggressively compete to maximize its award of this funding.”

A chart on Page 43 of the 2022 Draft Business Plan indicates that currently available funds total $11.4 billion, with $8.6 billion spent by October 2021 and $2.9 billion remaining. Total authorized funds to date add up to $25.2 billion, 24% of the high-cost estimate of $105 billion. Beyond the $11.4 billion mentioned before, the total includes $0.9 billion federal and $4.2 billion from future Proposition 1A (state) money. Beyond that, the rest would come from cap-and-trade funds, which bear some risk. The hoped-for amount is $8.8 billion, but the low estimate is $4.5 billion; 51% of that.

The same document contains a map (at 71) that gives a segment-by-segment breakdown of currently-estimated costs for Phase 1 of the project, from San Francisco to Anaheim. It adds up to $75.9 billion; a relatively-hopeful number which is 73% of the “high” estimate of $105 billion. A footnote mentioned additional statewide funding Caltrain Electrification, San Mateo Grade Separation, Rosecrans-Marquardt Grade Separation, and Los Angeles Union Station, totaling $1.298 billion. Another footnote revealed that the cost of building the Madera to Poplar Avenue segment had been reduced from $15.2 to $13.9 billion for single-track on the Merced-Bakersfield segment. Adding a second track would cost an additional $1.1 billion. Since the marginal cost of the double-track line is only 8% higher than the cost of a single-track line (presumably including passing sidings), it seems reasonable to spend the extra money. It also appears that, on a line where scheduling is so critical, adding scheduling flexibility in case of any difficulty (no matter how minor) and eliminating any resulting choke points would be a worthwhile investment, since a minor glitch could disrupt scheduling for a substantial period of time.

The overarching question about funding remains. It stands to reason that management would not promulgate a business plan that did not have numbers with which tit feels comfortable. Yet, the anticipated costs do not account for a large amount of risk, and we are living in very risky times. Everybody is hopeful that the worst of the death and disruption caused by the COVID-19 virus is behind us, but nobody can be sure of that. Inflation is the worst it has been in 40 years, and everybody hopes that the situation will be short-lasting. Again, nobody knows for sure, and planning on year-of-expenditure (YOE) dollars presupposes a certain interest rate or cost of capital to calculate the net present value of every transaction, even if it is given in YOE dollars. In other words, if inflation spirals out of control, those dollars will be very expensive by the time the YOE comes around.

So, while there is reason to believe that the managers are competent and working with the best available information, anything can happen. Maybe they will finish Phase 1 of the line, and maybe they will run out of money and have to live with what they have (as will their riders). If they complete Phase 1 and operate it, the same question and hope could be expressed about the Sacramento and San Diego extensions known collectively as Phase 2. 

HSR to Bakersfield?

There is often a difference between an insider’s and outsider’s perception of a plan. The Authority’s decision to start construction in the Central Valley is a case in point.

Simerly told me about the basis for that decision: “For one, starting in the Central Valley was a condition of federal seed money for the project, to invest in an historically disadvantaged part of California during a recession, with some of the worst air quality in the nation. But also, for two, it is the only place currently planned or under construction anywhere in the entire country where we could test our trains at voter-mandated speeds in excess of 200 mph. Such testing areas require flatter and straighter right-of-way, which is much more expensive in denser areas of the state. Purposeful efficiencies like this are the proof that we’re pursuing the system voters mandated.” He also included this link for “still or animated renderings, or photos or videos about the project or of construction”:  https://hsra.box.com/s/vyvjv9hckwl1dk603ju15u07fdfir2q8.

From the perspective of project managers, the stated reasons make sense, although they raise questions of why the feds required the Authority to build in that particular location. The idea of investing in a “historically disadvantaged part of California during a recession, with some of the worst air quality in the nation” may make sense to politicians, but not necessarily to folks who believe in the project because they want service. The “economic disadvantage” and “recession” arguments appear to relate back to the 1930s, when President Franklin D. Roosevelt’s “New Deal” programs provided work to millions of otherwise-unemployed workers and built a lot of public projects that are still in use today. While California may be one of the states where the polity is prepared to accept such a program, it is unclear that it would go over in a place where Democrats are not so firmly in control of state politics. The major problem with the “make work” approach is that, if the core mission of the project is to increase mobility for motorists and non-motorists alike, a “make work” project in one place can detract from starting construction in another place where people can benefit from more utility, sooner. 

As far as the environment is concerned, trains are good, because they provide an alternative to the massive energy consumption and harmful by-products from the typically American dependency on individual vehicles, no matter how they are powered. Then the question become whether it is better for the environment to build a new railroad next to an existing railroad, or to build a new railroad where there is no existing line that has been deemed unsuitable for passenger trains, especially ones running at high speed. Thus the question becomes: Where is the optimal place to build first?

For three days in April 2011, Amtrak 14, the southbound Coast Starlight, was re-routed over the historic Southern Pacific (now Union Pacific) route, on account of track work on the Coast Line, where it usually runs. Because of the detour, the San Joaquin Daylight lived again for three days. It was an adventure getting to Sacramento to ride it, but that trip demonstrated that, while it is scenic, the historic Tehachapi Pass route is slow. To this day, advocates debate whether it has been wise for the Authority to build near the existing San Joaquin route where conventional trains run several times a day, rather than between Bakersfield and Los Angeles. South of Lancaster, the railroad is relatively fast, due to the presence of Metrolink trains. Between Bakersfield and Lancaster, the railroad is so slow that everybody is required to get off the train and take a bus to the City of Angels. A bus bridge to Metrolink is not part of the trip.

The Authority’s plan has drawn harsh criticism, even ridicule, from some rider-advocates, but more from the general public and opinion leaders in the media. Not knowing the technicalities, they wonder why their state is spending so much money to build a railroad next to one that has several trains on it every day, rather than working toward the much-needed utility of a one-seat ride between the Central Valley (even if that ride is on a conventional train) and Los Angeles. They have been asking why build a faster railroad in the middle of the state, when they might still have to get on a bus at Bakersfield to continue their trip. It’s a good question. 

Let’s consider the actual time savings for a trip between San Francisco (or even the East Bay) and L.A. Union Station. The “worse case” scenario would be if the portion of the line in the Central Valley is completed, but then the money runs out. The time-saving of high-speed over conventional rail, which requires 180 minutes (assuming nothing is done to speed up conventional service on the current San Joaquin line) is 100 minutes, but any additional transfer time must be deducted from that, so the practical time saving is 90 minutes. At a currently estimated cost of $17.5 billion, that’s a lot of money for so little saving in time.

A more-positive scenario is that there is enough money available to electrify the Caltrain line and complete the segment through the Pacheco Pass to the wye, so trains can travel through Merced to Bakersfield. Using the mileage for segments listed on the map on page 71 of the Draft 2022 business plan, that distance is 297 miles. With the line completed between San Francisco and Bakersfield, it would take in the range of 120 to 135 minutes to go from end to end. It takes slightly over 360 minutes (six hours) to ride between Emeryville and Bakersfield on today’s San Joaquin trains, so the time saving for the trip to Bakersfield would be approximately 240 minutes (four hours). The time for the transfer to or from a bus at Bakersfield and the ride to Los Angeles would not change, and Amtrak now schedules almost exactly 150 minutes (2½ hours) for it. So the time between San Francisco and Los Angeles is now 510 minutes (8½) hours with the bus segment, and the time on a high-speed train and a similar bus would be 270 to 285 minutes (4½ to 4¾ hours), at least 90 minutes longer than the Authority’s goal of three hours, and with the bus segment still required.

While I remain uncertain about whether all of Phase 1 will ultimately be fully-funded, it appears that the current Caltrain line will be electrified in any event. Simerly said: “The project is now expected to be completed in 2024 when modern electric trains will whisk passengers down the Peninsula, providing a preview of what electric high-speed rail service will look like in the future.”

This raises the broader question of whether more riders could be attracted to an enhanced Coast Line, which could make the trip between San Francisco and Los Angeles in 390 minutes (6½ hours), assuming that could be done. Similarly, could a faster, but not high-speed, line possibly Class 6 track with a maximum operating speed of 110 mph or Class 7 for 125 mph, be built between Bakersfield and Lancaster to provide through service on an upgraded San Joaquin line to allow a shorter travel time? That scenario is beyond this article’s purview, especially since we don’t know how much those improvements would cost. Some thorough cost-benefit analysis would probably be useful, though, if there is ultimately only enough money to build “higher-speed rail” (HrSR) as an alternative to the more-expensive, but faster, true HSR.

Simerly also pointed to the fact that the high-speed track in the Central Valley would be the first railway line in the U.S. to accommodate trains operating at 200 mph. That would certainly confer a distinction on the line, but does the distinction mean much if it goes from one relatively-small town to another, without being part of a through-running route between California’s two largest metropolitan areas? It is not even clear how much use such a segment would serve as a “test track.” Without robust service, it has little or no such utility. If it has only light service, due to its isolation from other rail lines, it makes a good test track, but it does little to move people between places. This makes it all the more imperative that the Authority find the money for the entire line.

Summing Up the Past Three Years

I asked Simerly how things have changed for the project in the past three years. “We’ve known for a long time that we proceeded with our current construction contracts out of sync with our right-of-way acquisition, which has created delays,” he said. “In fact, we had little choice, given the federal government mandated where and when to start construction based on seed grant money. So, we’ve been planning for a while to implement a Staged Project Delivery process that would better sequence and streamline our processes, and we expect will lessen risk on new contracts. This process is exactly what’s now under way with these contracts before the Board. You can read about this process in our Draft 2022 Business Plan starting on page 9, among other places.”

Simerly also mentioned property acquisition, which could have caused a severe problem for Brightline in Florida, and is expected to doom the Texas Central project, which proposes HSR between Dallas and the Houston area. “Property acquisition is certainly a challenge we’ve been working through, but we have been delivering an often record number of parcels to the contractor per month,” he noted. “In fact, for the 119 miles under way, more than 90% of the more-than 2,000 parcels have been delivered to date.” Given the vital importance of property acquisition as part of preparing to build the line, that is a positive step.

Simerly offered this summary of where the California HSR project stands today: “We’re at an exciting moment in delivering the nation’s first 200-mph-plus, 100% renewably powered high-speed rail system. We have a reengaged and visionary federal partner supporting the promise of this project. We’ve reformed our organization and processes to better account for risks, and we have 119 miles and dozens of structures completed or under way in California’s Central Valley, creating 7,300 construction jobs and an estimated $13 billion in economic impact to date. No state is advancing high-speed rail at California’s pace. The opportunity is now to build the transportation infrastructure of tomorrow.” 

If nothing else, that statement shows determination. Certainly, if it is built, it will mark a new page in the history of North American passenger railroading. There are still challenges and risk, though, and the stakes are high. If the line is completed and ridership levels are high enough, the high-speed train will knock the airlines out of the Los Angeles – San Francisco market (and other nearby city pairs), as has been done in Europe, and demonstrate that HSR has a future. Deciding where such lines can be built and spending the money required to build them will be a different matter, always fraught with politics and dependent on hopeful economic times of long duration. The U.S. will not become like Europe, Japan, China and many other regions around the globe, where HSR is part of an overall mobility system that also includes conventional long-distance trains, corridors and local rail transit of all sorts, unless the political will and federal funding for a transformation away from automobile dependency take hold. This does not exist here yet. Still, the California experiment could establish HSR here to some extent.

If the current project is not completed because there is ultimately not enough funding available to complete it, and especially if there is not enough to build the spine between San Francisco and Los Angeles, a line ending in Bakersfield will continue to serve as an object of scorn and false narratives about how the U.S. does not have the population density to support true high-speed rail. As things look from here, the California HSR Project is facing yellow signals, and it could go either way.

David Peter Alan is one of America’s most experienced transit users and advocates, having ridden every rail transit line in the U.S., and most Canadian systems. He has also ridden the entire Amtrak network and most of the routes on VIA Rail. His advocacy on the national scene focuses on the Rail Users’ Network (RUN), where he has been a Board member since 2005. Locally in New Jersey, he served as Chair of the Lackawanna Coalition for 21 years, and remains a member. He is also a member of NJ Transit’s Senior Citizens and Disabled Residents Transportation Advisory Committee (SCDRTAC). When not writing or traveling, he practices law in the fields of Intellectual Property (Patents, Trademarks and Copyright) and business law. Any opinions expressed here are his own.

Tags: , ,