A billion here, a billion there:
California bullet train officials on Tuesday released a new draft project blueprint that acknowledges that costs have risen roughly $5 billion but seeks to address several issues that have generated blowback.
The 2022 business plan estimates that the full, 500-mile, high-speed system between Los Angeles and San Francisco will cost as much as $105 billion, up from $100 billion two years ago. In 2008, when voters approved a bond to help build the railroad, the authority estimated that the system would cost $33 billion.
At this rate, it will end up costing about $200 billion by the time it's finally finished in 2040 or so. Assuming it's ever finished, of course, which is pretty unlikely at this point. By 2030 we'll have a nice, fast train from Bakersfield to Merced that no one will use, and that will be about it. Hooray for California.
Article is behind the LA Times paywall.
What's driving these costs so high so fast? They haven't even done anything yet.
I do think a SD to LA to SF train would be a great thing to have, but not all such things are practical.
“Some of the higher costs undoubtedly stem from changes the authority was forced to make to avoid impacts on San Joaquin Valley communities.”
If you have an Apple device and use Safari you can get past the paywall by using reader view. I’m told other browsers have this feature also.
I voted for a train from San Francisco to Los Angeles, not a train from Merced to Bakersfield. I don’t know the demographics of the vote at the time, but I suspect the people who live in the counties the train would serve voted against it.
There were three stages for the towns in the Valley:
1. Oh, we want/ need/ demand a stations in our town! Count us in.
2. Wait, you mean there'll be tracks and noise and things?
3. Stop the destruction of our farmland! Stop the tyranny of the train mongers!
This stretched out over several years.
This is like nuclear power: You can't simultaneously initiate lawsuits for the express purpose of causing cost overruns (construction delays will definitely do that at the far end), and then complain about the high cost.
And yet that's exactly what's happening to the local rail in Santa Cruz.
"They haven't shipped any cargo in five year!"
"...but you didn't repair the trestle, so no cargo could be shipped..."
The agency ridiculously low-balled estimates of what it would cost to acquire land and right-of-ways in the Central Valley. Then there were an number of unanticipated (whether they should have been is another question) engineering/infrastructure problems, such as the electricity grid in many parts of the valley not being robust enough to power the train. That's just for starters...
Kevin,
Are you the same Kevin Drum who (correctly) hammers away at inflation adjustment when discussing multi-year costs, etc.? Are these figures adjusted??
Why would they be? This isn't comparing the price of identical things at different points in time. It's comparing the exact same thing from the point of approval to the point of completion. Any extent to which you would lower the cost of construction for inflation adjusted dollars, you also have to adjust for the fact that a train that doesn't run until many years after it was first projected has also provided less value.
The reason for considering inflation is simple. A dollar in 2008 is worth $129 today. https://www.usinflationcalculator.com/ Inflation erodes dollar value, even at the very low rates we've enjoyed until last year.
I had to look up using your link because that sounded crazy, and you are only off by 2 orders of magnitude (aka 10^2 or 100). $1 in 2008 is worth $1.29 now. That is still a 29% increase, but thadoesn't come close to making up for the gap since we are north of 200% of the original estimate and counting.
I am all in favor of high speed rail, but this has clearly bee a boondoggle for a very long time and should just be scrapped at this point.
...and...I can't edit out my typos and poor grammar because WordPress sucks...a lot.
Obama really was American Mugabe.
But it's true: you can't spell Hyde Park without Harare.
Sorry but these are projected costs. The money has not been spent. They will rise with inflation until the project is carried out and the land is actually purchased and the work done at wages then in effect. This happens to every project that lasts longer than a few months.
Kevin, the inflation scold is really disingenuous to skip it in exactly this example just because this one thing has been a bee in his bonnet since it started.
I can see that this project has gotten into a cut de sac and probably should be dropped to avoid losing more money. But the fact that that happened reflects very bad on California (or the US ). Any competent government would work on a line from Vancouver to San Diego
You would never want to HSR from Vancouver to San Diego. The segment from Southern Oregon to Sacramento would have almost no ridership.
What about Greater Idaho residents making pilgrimage to QARON Rodgers's birthplace to do obeisance to ANTIVAXXXIA?
Sorry but these are projected costs. The money has not been spent. They will rise with inflation until the project is carried out and the land is actually purchased and the work done at wages then in effect. This happens to every project that lasts longer than a few months.
Kevin, the inflation scold is really disingenuous to skip it in exactly this example just because this one thing has been a bee in his bonnet since it started.
I can see that this project has gotten into a cut de sac and probably should be dropped to avoid losing more money. But the fact that that happened reflects very bad on California (or the US ). Any competent government would work on a line from Vancouver to San Diego--and if it were really competent the line would now be in operation
Kevin keeps me from coasting ... which, for my age, is a good thing. 'Nuff said.
As per your earlier post, we can do great things. This is the exception that proves the rule.....
Curious, what percent of the system will actually be run at "bullet train" speeds?
The part being built.
If we had a normal political system, rather than the one we have which offers up the absurd Larry Elder as the alternative, this would be a great issue for the opposition party.
Anyway, take all that money and put it into additional buses, plus making public transport free, in LA, and you’ll really have done something.
How???? There is no bullet train and never will be. It's politically dead and always has been.
The bottom line is America's first HSR project shouldn't have been LA to SF.* The technical issues are nearly insurmountable. Even the Chinese don't build HSR in places where it's technologically infeasible.
*I've seen enough analysis to believe Central LA to Central SF would work in terms of passenger numbers and economics; but nature obviously doesn't care about our need to make the sums add up.
Should have been Chicago to St. Louis.
I do think in principle HSR connecting the heart of the SF Bay Area to the heart of metro LA could make sense if the engineering could be worked out in economical fashion. But that's not where this project is currently at, because of that mountain pass to get out of LA County (name escapes me).
I'd say the obvious slam dunk HSR project in the US would be Bos-Wash.
Tehachapi Mountains
The problem with DC to Boston is that the area is so built up that it will require a ton of property takings of very expensive land. I hope they do figure that out someday but for now they are planning HSR from DC to Richmond and possibly DC to Baltimore.
And the part that isn't built up (i.e., Connecticut) is inhabited by rich, well-connected people who don't want HSR running through their quaint little towns (or their country estates). There will be no HSR between Boston and NY, that's for sure.
Baltimore to Charlotte: connect the Queen Cities.
As a (transplanted) Chicagoan, Hear, Hear! Somewhat more objectively, it's all flatland and -- more important -- acquiring right-of-way is way cheaper.
What "insurmountable" issues are there, exactly? Spain has managed to build a large system (Europe's largest) that connects most of its major cities and still did so at a good cost per kilometer in comparison to many other countries. It also has a fairly similar geography to CA. Italy is another example of an HSR system that has been built despite its very challenging topography.
Sure, building out the existing Acela line on the East Coast should obviously be a top priority, but connecting such major metropolitan areas as SF and LA (the fifth and second largest, respectively) with their pretty optimal distance (and several sensible stops along the way) should make them also good candidates for HSR.
Irony: The main contractor on California's rail is the Spanish one. But they under bid a bunch of things and ended up having to rebuild some sections and getting ripped off by local contractors.
The fact that something can be built one place does not prove it can be built somewhere else. In this case, it's extremely hard to cross the Tehachapis due to some unique geographical constraints.
"Geographical constraints" such as? Yes, a place can have a somewhat unique set of geological features, but the same geological elements recur in varying degrees all over the world. Tunnels have still been built despite such challenges.
It really isn't that simple. The basic problem is twofold:
1. The Tehachapis have a very steep north face with twisty canyons running through it, which makes it very hard for trains to make it up the hill; and
2. The Tehachapis were formed by an earthquake fault that periodically (once every 120 years or so, and it's overdue for the next one) throws off 8.0+ magnitude earthquakes.
(1) means you can't go up the hill without a ton of tunnels, and (2) creates a real problem building your tunnels (because you can't take the risk that the once in 120 year quake collapses a tunnel and kills 300 people riding a train).
This, by the way, is why there isn't any low speed passenger rail between LA and Bakersfield, and only one low speed through freight track (which goes very far out of the way adding hours to the trip rather than traveling straight), despite the huge profit potential of additional connections (because once you get over the hill, it's the fastest way to the San Francisco Bay Area). It's very very hard to engineer tracks over that hill.
And the CAHSR people haven't even tried to do it- they started on the flatland because they still don't know how they are going to get over the pass.
1) The steepness of the mountains is not an issue for HSR since it's designed to go through mountains and not over them, keeping climbs and descents very gradual. Yes, that means expensive tunneling, but that's not "insurmountable." These tunnels could be in use for a century or more so the initial high costs shouldn't be decisive.
2) Japan is one of the most seismically active regions of the planet and experiences relatively frequent, large earthquakes. That didn't stop them from putting roughly 30% of the Shinkansen under ground (or in certain cases, under water).
I just checked: Barcelona - Madrid is almost exactly the same distance as LA - SF. That train ride in Spain takes less than 3 hours! That's definitely muy bueono</iY.
It's not nature blocking us.
It is. The reason they are stuck mucking around in the San Joaquin Valley is because the most useful planned segment, LA-Bakersfield, is impossible to build at any cost that the taxpayers might pay. So they were hoping to build a "cheap" useless segment and then demand the money (probably multiples of $100 billion) to build the useful segments. Of course they couldn't even get the cheap segment done so that strategy didn't work.
Umm...pretty sure Japan is very seismically active,mountainous, and densely populated, if those are some of the "technical issues" you think exist in CA. Because other than that I can't think of any significant technical issues that would derail (cue rimshot) this project.
Again, building something in one place does not prove you can build in another.
This whole business is so un-businesslike that it evokes only laughter when I read about it.
So now we have a predicted cost overrun that Kevin estimates will escalate another 90% in the next 18 years. For a guy who seems to have a pretty good handle on numbers, Kevin has not noticed how unrealistically optimistic his estimate really is. His 90% growth in 18 years is about 3.6% per year, compounded over those 18 years. Really???
Suppose we assumed 5% per year? That would lead to an 18 year estimate of $250 billion. Or what about 6% per year? That would give an 18 year estimate of $300 billion. Amazing, isn't it, how compounding works!
A rule I taught my project managers for 20 years:
(a) For short-term projects, assume the worst possible scenario. Now do your estimate based on that assumption. Call the result your normal-case estimate, and build in some more slack for the unanticipated problems.
(b) For long-term projects, follow the steps of (a), then take that result as your best-case estimate, and redo your estimate for a worst-case, based on the above being your best-case.
For politics, it's more about getting the camel's nose into the tent...
Look at every defense department estimates of cost for weapons systems.
When this was passed I said something to the effect of, and I was wrong, "that it will cost twice as much and go half as far".
It's too bad we don't have a viable alternative to vote for in this state. The alternatives are all bat-bleep crazy.
On the good side, single payer health care died, this year. I'm not against the concept, I just don't trust the people putting the plan together.
If you're building a British-style public healthcare system (largely) from the ground up, or offering a mixed system of subsidized private insurance like Obamacare, it's doable. If you're trying to take a massive private healthcare system public, it's always going to be a clusterfuck. Vermont, which is like 2% the size of California and not nearly as diverse, tried it and found it was simply too expensive to fund w/o raising taxes substantially, which was politically infeasible, even in relatively liberal Vermont. It would be like the bullet train, but with health care. Except if you pull the plug on the bullet train, nobody dies.
They should have had Jane O'Meara Sanders develop the same funding mechanism as she used for Burlington College.
This.
I'd very much like for the US to have a developed-world universal health care system, but I had zero confidence in the California plan and was happy when it died.
The train debacle was front of mind. I heard about this plan and thought "It's gonna cost three times what they say, arrive years late, and cover only people in Kern county. Hard pass."
Because it would need to interact with Medicare and Medicaid, a state level single-payer plan is unfeasible, no matter who is putting it together. It needs to be a federal program.
This train project is a living example why folks lose trust in Government. Confidence in government has declined from the high 70s during LBJ to the low 20s today. I THINK a lack of trust in government is a consequential challenge, sub rosa, to big Democrat plans such as single payer health care and even Build Back Better legislation.
https://www.pewresearch.org/politics/2021/05/17/public-trust-in-government-1958-2021/
What, Gavin Newsom was going to use high-speed rail to relocate Blacks to previously Okie-heavy areas in the name of desegregation?
It really is. It greatly reduced my confidence in California's government. I'd never trust it to execute on a project of this size again.
As a lifelong liberal Democrat now converted to the GQP over California HSR overruns, are you also quite concerned about Chappaquiddick?
Wow, it's as bad as the transcontinental railroad with its immense overruns and taxpayer ripoffs. That was supposedly privately funded, but it relied on such massive federal subsidies that people were still bitching about it into the 1890s.
Since this is a generally optimistic blog, I'll note that transportation infrastructure often induces demand. Where are the warehouses for the SF & LA areas located? Could this be the middle mile? Isn't this agricultural heartland? There's also the matter of the upgraded electrical power system. Would that have knock on benefits for solar power distribution?
Even if they pull the plug, they'll still have the assembled right-of-way, and I'll bet that in the long run it will be used for something more than a bike path.
You can't induce demand if it is impossible to build.
Bottom line, this train must either reach LA or San Francisco/Oakland, or the project is an obvious failure.
The biggest value-added would be Los Angeles, because except for the coastal train, you cannot take a direct train from Los Angeles north, and the coastal train completely bypasses the San Joaquin Valley. That would also the most challenging and expensive part, but would be brilliant in linking up the state.
The easiest part is obviously San Francisco/Oakland, since we already have Amtrak service between there and Bakersfield (before you have to get on a bus to reach LA and parts further). If they end up being unable to get even this stretch of track done and in service, they never should have even started this project, considering LA is the hardest and most expensive part.
Aside from that, I don't know much this has to do with legalities like acquiring easements and all of that, which is tedious and time-consuming, or if the original estimates were just gross incompetence.
It's not, because the rails there are so out of date. Local nimbys have blocked any adaption to increase speeds, and it was tough enough to get it to have welded rails a decade ago.
In SF, the train uses 'the old tunnel' aka the one that didn't fail in 1906. It got overhauled, finally, but it's still less optimal. And the line to Oakland north and south didn't get alot of love or updates.
It happens that the most needed segment, which you correctly identified as LA-Bakersfield, is also the one that nobody knows how to build at feasible cost.
How much has been wasted/spent so far?
But that's barely inflation, Kevin. C'mon.
Any time there’s a public works project, particularly a large one, there’s always a hue and cry about how badly managed the project must have been because costs escalated well beyond initial estimates.
Anyone who participates in this Kabuki play doesn’t understand the rules. Public works projects are ALWAYS lowballed. Because if they weren’t, the hue and cry would be up front and the project would never get off the ground.
The insiders know how the game is played. The politicians know full well that the estimates they are given are low, and they play along, telling the public those low figures. The bidders know they have to play along and bid low, knowing full well that “change orders” will bring the price up to a level where they actually make a profit.
Then, when the revised estimates come in, there are large sunk costs and a bunch of inertia. So these projects get built and are never abandoned.
Rinse, repeat.
Bear in mind this was a voter initiative. It's one thing to lowball a legislature who knows how the game is played. But you pay a huge cost of credibility when you lowball the voters- who were told they could have the whole thing for less than $10 billion.
"By 2030 we'll have a nice, fast train from Bakersfield to Merced that no one will use..."
And how are those self-driving cars going ? Fleets of 'em by the end if this year ? Uber on a massive earnings upswing ?
WAYMO BE THERE.
At what point does it make sense to cancel this project?
The only portion that still makes sense is the upgrade of the Peninsula/Caltrain line between San Jose/Silicon Valley and San Francisco to electric and higher speeds. That's both doable and would be enormously beneficial once completed.
The rest of the project should be killed.
Monoraaaaail . .
MonoRAAAAILLL. . . .
MONORAAAAAIIIILLLLLL!!!