California HSR

[GDRRiley]

He said this toward the end of his rant...

He's not investigating it to fix it, he's investigating it to kill it for good, just like he tried to do the last time, and this time he'll succeed.

There is nothing he can do to kill it, even if he delays all recent federal funding that won't kill the project and its likely CA can move some money around now to keep the project on track, that can then get used for something else in 2-3 years

 

[cirdan]

This is sadly true. This mess has likely killed any hope for true high speed rail in the US.

One could hope that the mistakes can be analyzed and that future projects can learn from them. This way the losses would not be totally in vain.

I guess one can dream.

 

[frequentflyer]

This is sadly true. This mess has likely killed any hope for true high speed rail in the US.

The original cost of the California High-Speed Rail (CAHSR) under Obama was $33 billion. Now, it is estimated at $128 billion. Someone grossly underestimated the project, did not know how to design a high-speed rail project, or was lying.

The reasons should be investigated. Why should a voter trust any "grand" scheme again?

 

[Neil Rubin]

The original cost of the California High-Speed Rail (CAHSR) under Obama was $33 billion. Now, it is estimated at $128 billion. Someone grossly underestimated the project, did not know how to design a high-speed rail project, or was lying.

The reasons should be investigated. Why should a voter trust any "grand" scheme again?

The text of Proposition 1A that California voters approved in November 2008 (before Obama took office) estimated a cost for the whole system of $45 billion (https://lao.ca.gov/ballot/2008/1a_11_2008.aspx). The 2024 business plan has a Phase I (S.F. to L.A.) cost of $89–128 billion. You can cherry pick the $128 billion number, but it is meant to include quite a large contingency for unexpected costs. Nonetheless, the actual costs are clearly higher than estimated. As I understand things, there are a few big reasons:

(1) Much higher construction cost inflation than forecast, both because we've had more inflation than expected (particularly between 2020 and the present, and particularly in construction) and because construction has been much slower than forecast because the private and federal funding the state was hoping for didn't come in the amounts hoped for. Based upon the data at https://edzarenski.com/2024/01/17/construction-inflation-2024/, "non-building infrastructure" inflation has been about 65% since 2008. Highway cost inflation has been 100% since 2008.

(2) There was money wasted early on because the conditions of federal grants pushed the state to start work before it was really prepared. Eminent domain and utility issues slowed down construction work and drove up costs. Not good, but not a primary driver of the topline cost increase.

(3) What is actually getting built is bigger in scope than what the estimates were based upon, in terms of aerial structures, grade separations, freight facility relocations, noise mitigations, and the like because BNSF, UP, and cities along the route (or combinations of these: city demands route than crosses freight route, and freight railroad demands that there be zero—or beneficial—impact on its operations) demanded these expensive changes, and the HSR Authority didn't or couldn't push back.

There is also no doubt a lot of money that could have been saved if the state hadn't been so reliant on outside private consultants, who are expensive and don't really have strong incentives to control costs. There are also probably excessively generous deals with unions and excessive insistence on Buy America rules.

But I think the real failure here is not incompetence by the folks at the Authority or their consultants. For them to have succeeded, they needed a predictable and realistic funding source and political backing that enabled it to say no sometimes to the demands of local governments and the freight railroads. Instead, we've had a state government unwilling to kill the project but only willing to give it the minimum funding to keep it alive and an Authority that politically can't afford to make enemies, even if that's what it needs to be willing to do to control costs.

 

[PaTrainFan]

The original cost of the California High-Speed Rail (CAHSR) under Obama was $33 billion. Now, it is estimated at $128 billion. Someone grossly underestimated the project, did not know how to design a high-speed rail project, or was lying.

The reasons should be investigated. Why should a voter trust any "grand" scheme again?

This is one example of egregious "low balling" of public infrastrucrure projects. However it happened, intentionally or unintentionally, this is one example of horrendous cost "low balling" of public infrastructure. It happens fairly routinely with highways and military projects. Estimates, for whatever reason, are typically made to be palatable in the beginning then balloon to a point where the projects become too big to fail, thus abandoning is seen as unthinkable. I cannot fathom what would happen here if the plug is pulled.

 

[GDRRiley]

The original cost of the California High-Speed Rail (CAHSR) under Obama was $33 billion. Now, it is estimated at $128 billion. Someone grossly underestimated the project, did not know how to design a high-speed rail project, or was lying.

33B number isn't really rooted in anything other than some very early estimates using European costs. By the time they started looking at the actual designs costs increased to nearly the same amount as it is now, They also at some point changed to year of expenditure which didn't help the number

 

[cirdan]

I cannot fathom what would happen here if the plug is pulled.

I don't want to be advocating such a solution. It would be a huge tragedy.

But wasn't the strategy from day one to build HSR in sections in such a way that each section built has standalone utility. In other words, the built sections would not rot away but would offer worthwhile improvements to corridor services.

 

[Neil Rubin]

33B number isn't really rooted in anything other than some very early estimates using European costs. By the time they started looking at the actual designs costs increased to nearly the same amount as it is now, They also at some point changed to year of expenditure which didn't help the number

I’d really like to understand the history of this fiasco of growing cost estimates better. Do you know of reliable, accessible sources that explain the early cost estimates and how they changed with time? Were the 2006 or 2008 estimates really all using then-year costs, with no accounting for inflation? Aside from changing to year of expenditure and the effects of inflation, where is the biggest change in the estimates over the past 20 years? For example, do changes in the costs of the mountain range crossings account for a large part of it, or is it a whole bunch of things? What are the biggest differences between those European projects and what CAHSR plans to build that makes the European costs unrepresentative? There are just general differences in cost for infrastructure construction in different places, but are there different engineering decisions that drive the cost difference, e.g., the 220 mph speed or a lower maximum grade, driving up the lengths of tunnels, bridges, and viaducts? I’m sure all of the answers are out there to be found by someone willing to do the legwork, but alas there are other priorities to spend time on.

 

[Neil Rubin]

I don't want to be advocating such a solution. It would be a huge tragedy.

But wasn't the strategy from day one to build HSR in sections in such a way that each section built has standalone utility. In other words, the built sections would not rot away but would offer worthwhile improvements to corridor services.

You could have gone with a purely “bookends” focused strategy, but things with real benefits to non-HSR services, like Caltrain electrification, L.A. Union Station through-running, and L.A.-to-Anaheim LOSSAN improvements are happening. They’ve been much too slow and at least in the Caltrain case too expensive, but that’s more on the local folks actually building these things than an HSR problem, per se.

You could also have just put the HSR goal to the side and spent more money on LOSSAN, Capitol Corridor, ACE, etc. infrastructure that will never been used for HSR.

But outside of the bookends, where do you build HSR infrastructure with worthwhile benefits to corridor services? The Central Valley segment does actually cut two hours of travel time from the San Joaquins and provide comfortable trains and lovely stations for people who currently travel on that corridor. Punching tunnels through mountain ranges or even upgrading San Jose to Gilroy doesn’t get you much utility unless you’ve built large parts of the rest of the system already.

If you knew Brightline West was coming in the current timeframe, there would have been a case for prioritizing L.A. to Palmdale over the Central Valley, in order to get a real HSR service from L.A. to Las Vegas.

 

[cirdan]

It would be wrong to assume that cost overruns don't occur in Europe. For example the Swiss base tunnels were initially costed at 8 billion, with costs rising to 23 billion, even though the project was scaled back (not all parts were built, but are still being kept open as options for future expansion)

 

[OmahaRoad]

The text of Proposition 1A that California voters approved in November 2008 (before Obama took office) estimated a cost for the whole system of $45 billion (https://lao.ca.gov/ballot/2008/1a_11_2008.aspx). The 2024 business plan has a Phase I (S.F. to L.A.) cost of $89–128 billion. You can cherry pick the $128 billion number, but it is meant to include quite a large contingency for unexpected costs. Nonetheless, the actual costs are clearly higher than estimated. As I understand things, there are a few big reasons:

(1) Much higher construction cost inflation than forecast, both because we've had more inflation than expected (particularly between 2020 and the present, and particularly in construction) and because construction has been much slower than forecast because the private and federal funding the state was hoping for didn't come in the amounts hoped for. Based upon the data at https://edzarenski.com/2024/01/17/construction-inflation-2024/, "non-building infrastructure" inflation has been about 65% since 2008. Highway cost inflation has been 100% since 2008.

(2) There was money wasted early on because the conditions of federal grants pushed the state to start work before it was really prepared. Eminent domain and utility issues slowed down construction work and drove up costs. Not good, but not a primary driver of the topline cost increase.

(3) What is actually getting built is bigger in scope than what the estimates were based upon, in terms of aerial structures, grade separations, freight facility relocations, noise mitigations, and the like because BNSF, UP, and cities along the route (or combinations of these: city demands route than crosses freight route, and freight railroad demands that there be zero—or beneficial—impact on its operations) demanded these expensive changes, and the HSR Authority didn't or couldn't push back.

There is also no doubt a lot of money that could have been saved if the state hadn't been so reliant on outside private consultants, who are expensive and don't really have strong incentives to control costs. There are also probably excessively generous deals with unions and excessive insistence on Buy America rules.

But I think the real failure here is not incompetence by the folks at the Authority or their consultants. For them to have succeeded, they needed a predictable and realistic funding source and political backing that enabled it to say no sometimes to the demands of local governments and the freight railroads. Instead, we've had a state government unwilling to kill the project but only willing to give it the minimum funding to keep it alive and an Authority that politically can't afford to make enemies, even if that's what it needs to be willing to do to control costs.

Neil, this is the best and most succinct analysis I've ever read of what actually has caused the 'cost overruns' on CAHSR. By far!

Would you possibly be ok with me saving this explanation and possibly sharing the text or parts of the text of your response in the future? It would be very helpful ammunition to have at the ready when I see people spouting utter lies about CAHSR costs and the like. It would of course be anonymous but I would be happy to attribute to you if desired. I totally understand if you aren't comfortable with that, but wanted to ask because your writing is so clear and helpful.

 

[MARC Rider]

The reasons should be investigated. Why should a voter trust any "grand" scheme again?

The voters have themselves to blame. Apparently, the text of the ballot imitative specified that the system was to have a 220 mph maximum speed, a specified maximum travel time that was probably overly optimistic, and even what stations they should build. That means that the HSR Authority doesn't have a lot of flexibility in terms of cutting costs. Added to that, various NIMBY activities held things up and forced construction decisions that added to the expenses.

And investigation to determine why the expenses ballooned might be a good idea if the point is to learn from it and apply the lessons learned to future projects, but I'm afraid these calls are mostly to advocate punitive measures on people who probably don't deserve it and prevent the spending of public funds on rail infrastructure projects in the future. The only reason one would need a punitive type of investigation if there were allegations of actual corruption, where money was diverted from the project into somebody's pockets, and I haven't been hearing those sorts of allegations about this project. I think that the project managers may have made mistakes, but they were forced in that direction by the actual wording of the ballot initiative. If you want to point fingers, I guess one should point at the pro-rail activists who drafted a ballot initiative with unrealistic goals and the majority of the voters who approved it.

Come to think of it, maybe they should have a new ballot initiative authorized the state to build the line giving more flexibility about top speed and travel time, etc. to allow the state to cut costs and integrate better with the non-HSR intercity and commuter rail networks.

 

[MARC Rider]

They had a panel discussion at this year's TRB about "International Standard High Speed Rail in the US" that was mostly about the California High speed rail project. From what was said, it looks like the Merced - Bakersfield line is very close to being finished. The real problem is getting across the mountains on either end, but the California, but the Secretary of Transportation of California, who was on the panel, seemed pretty optimistic that they would be able to do that at a reasonable cost.

What I'm interested in is how they're going to operate the trains when this stretch is finished. Mr. Omishakin, the Secretary of Transportation did point out that the population of the Central Valley, between Merced and Bakersfield, is fairly significant, so I suppose that Merced - Bakersfield trains might have a decent ridership, and it might jump-start some urban development in what is a depressed part of the state. Otherwise, they could run the San Joaqins on the new line, thus providing a faster Oakland - Bakersfield ride. Unfortunately, the actual HSR rolling stock requires catenary, which isn't available between Merced and Oakland (or Sacramento), so it would look weird seeing a 220 mph HSR trainset being pulled by a couple of Chargers at considerably less speed.