Cost of California bullet train system rises to $77.3B, take 5 years more | Hacker News Reader

Cost of California bullet train system rises to $77.3B, take 5 years more

latimes.com

39 pointsslapshot33 comments

I know this project is costing a fortune but I'm a proponent of it. The connection of so many cities will eventually be a boon for California.

The red line in Los Angeles had a similar history with cost overruns and extreme costs but now it's a vital part of the Los Angeles transportation system. I see the same happening to this system.

We marvel at the Roman aqueducts not because of their costs but at the fact that they were built in the first place. Similarly, we will say the same thing about the bullet train.

I wish there was a way to opt out. I will willingly sign a contract to never use said system (if it ever gets finished), in return for not sponsoring it anymore with my taxes.

I'd prefer those taxes to instead go somewhere useful, like paying teachers better.

It seems like even if you're not literally riding the train, you'll benefit from it if it has a positive impact on the economy, traffic, property values, etc.

But is that a greater benefit than the effects of not taxing away all that money from folks who’d prefer to spend or invest it differently? Those effects are difficult to appreciate because they’re so diffuse but could easily outweigh the more tangible benefits of what’s otherwise a boondoggle.

I mean, we all want our taxes to go to things, but being part of a democratic society means we don't get to dictate to that degree. Sometimes we do things to help people other than ourselves.

I don't think anyone is opposed to having a high-speed rail line. What people are opposed to is the insane cost, the schedule delays, and the poor output. If you compared this to high speed rail in Japan or China - it would come out looking horrible for California on every meaningful metric.

Regarding your point about Rome - I'm not an expert, but I'd bet the aqueducts provide a valuable service. What are your estimates for how many passengers would need to ride the rail in order to make it worthwhile, and your estimates for how long it would take to serve that many passengers?

Think about the huge amounts of money and the value of the rail (delayed by when you could get that value) and then compare that to other ways you could use that money - for example by leaving it in the hands of the taxpayers.

It’s important to remember that it is much easier to get property to build on in other countries. In China (which you cite) you don’t get a choice at all - your property is sold for whatever amount the appropriate gov agency believes it is worth.

The article says buying land for the track is a significant problem - I will happily bet that they are not paying the pre-rail market value of that land.

Actually, it’s not as bad as you think. The Shanghai-Beijing line had a final budget of around $35bn for an 810-mile service. California is looking to spend $78bn for a 520-mile service. That means a cost-per-mile of $150m for California vs. $43m for China - in other words, California’s system is around 3.5 times as expensive as China’s.

Honestly, given the difference in land values and regulation, that doesn’t seem entirely unreasonable.

It's actually better than that. The Shanghai-Beijing line cost ~$40 billion in 2018 dollars. The cost estimates for the California line are in year of expenditure dollars, which means the $78+ billion is amounts that are spent from 2008 until completion in 2033.

The dollars spent early in the project cost more because of inflation, and along the same lines, the dollars spent later in the project cost less because of inflation.

For instance, if we spend half of the $80 billion in the last 5 years of the project, that's something like $30 billion in 2018 dollars even though it's $40 billion in 2028 to 2033 dollars.

> but now it's a vital part of the Los Angeles transportation system

Losing its vitality, though http://www.latimes.com/local/lanow/la-me-metro-ridership-201...

It's expensive, but the LA Times article is leaving out a critical piece of information. The costs they're talking about are in year of construction costs, not 2008 or even 2018 dollars.

For instance, if the expected completion date is 2028 and the cost is $70 billion in 2028 dollars, any delay will increase the absolute cost because of inflation, even if the cost in 2028 dollars stays the same.

While $80+ billion in costs by 2033 is a lot of money, it's not actually twice the initial estimate of $40 billion in 2008 dollars (something like $55-70 billion in 2023-2033 dollars, assuming ~3% inflation/year).

Man this must be a hey-day for property lawyers in the valley.

You could just sorta guarantee this would happen -- the nimbyism in California is like full-tilt. I'd suspect the costs will go up even more from here and the time to completion will be even longer than this

It's a shame we can't have nice things here.

That's almost the cost of the $100 billion international space station.

And it's not even a maglev. A maglev would've been way cooler.

It's probably going to be more than the US contribution of $50 billion in 1994 ($85 billion in 2018 dollars), but I think it would have to exceed $200 billion in real dollars to match the total real cost of the ISS.

https://www.space.com/9435-international-space-station-worth...

Still, at > 2 times the real original estimated cost, it's a good example of why we should go high instead of low when it comes to estimates.

I'm also curious how much it would cost to build out the same capacity via airport construction/expansion. If we're at more than double the cost running rail lines through the central valley, imagine how much building another LAX, SFO, SAN, etc... would be.

Its not estimates that are high, its actualy costs that are high.

It's both estimates and costs. Mostly estimates at this point from what I understand, unless they already finished the line through the central valley (scheduled for 2025) and haven't told anyone yet.

https://www.mercurynews.com/2018/01/17/worst-case-scenario-h...

> Still, at > 2 times the real original estimated cost, it's a good example of why we should go high instead of low when it comes to estimates.

It's also a good example of why no one goes high when it comes to estimates.

The realized cost is now to a point where it'd be politically unpalatable to approve the project. But since it's already in process, approving the incremental funds is essentially inevitable and won't cause much political backlash.

These estimates aren't low by accident. They're generally as optimistic as possible while still being in the realm of marginally plausible (even if wildly improbable). Once the ball gets approved and rolling, the momentum alone is usually enough to hurdle past objections as estimates meet reality, while diffusing responsibility (and therefore repercussions) along the way.

After looking at it some more, there's a relatively simple explanation. The initial cost estimate was $40 billion in 2008 (voters approved $9.95 billion in bonds).

https://ballotpedia.org/California_Proposition_1A,_High-Spee...

The thing the LA Times and others are leaving out is that the CA high speed rail authority estimates are in year of expenditure dollars, not 2008 dollars or current dollars.

http://www.hsr.ca.gov/docs/about/business_plans/DRAFT_2016_B...

So what they're estimating is that by the time the line between socal and norcal is complete, around 2033, we'll have spent $77+ billion (up to $98 billion).

That's not $77-$98+ billion in 2008 dollars, it's $77-$98 billion in 2008 through 2033 dollars, depending on when those dollars are spent.

For instance, if we spend most of the money near the end of the project, and the total cost is ~$80 billion, that corresponds to $50+ billion in 2008 dollars. It's still higher than the initial estimate, but it's not nearly as high as going from $40 billion in 2008 dollars to $77+ billion in 2008 dollars.

Honestly, how long will it take for this to be a good investment? My guess is more than 50 years. Will we need a bullet train in 50 years? Probably not. Anyone that wants to go anywhere will be able to step inside a pod, take a nap, and wake up wherever they want to be. Such a waste of resources...

In other news, China will have about 28,000 miles of high-speed rail by 2030.

When I was in kindergarten, there was a proposed (Washington) metro station down the street. Metro put up a sign, "Coming soon...". Twenty years later the sign fell down. The station was eventually built. The first time I used it, I took my daughter, who was just about to start kindergarten. Now, I live in northern CA, just a mile from where the proposed bullet train will run. When people talk about the impact, I tell them, that's our children's problem, and reference the above story.

I'm all for this train, but if it costs this much to build and even more to maintain, and it's not even the newest train technologies, then it's probably just doomed to fail.

Just did some quick calculations, for the bullet train connecting beijing and shanghai which is about twice the length of the cal bullet train system, the chinese spent about $30B

Oddly that makes it sound not so bad; given the property and environmental issues and the high cost of labor I’d have expected a larger multiplier.

This reminds me of the aerospace industry adage (F-22 / F-35 related), that if you extrapolate the continuing growth curve of fifth generation air superiority aircraft acquisition costs, in 75 years from now the USAF will be able to afford to purchase a single fighter and it will cost $9.5 billion.

I would consider this project a success if they can build a functioning high-speed rail line from Fresno to San Jose. Connecting the Central Valley to Silicon Valley - or the fifth largest city to the third largest city in CA - would bring tremendous benefits. If this portion of the project can be proved out, building out to SF and even Sacramento is feasible. I'm not sure it's ever going to be economically feasible to tunnel through the mountains to SoCal.

tl;dr - Since in California all construction requires an absurd amount of licenses, approvals, environmental studies, etc., and because local residents can sue for almost anything, the cost has skyrocketed.

Isn't this what eminent domain was created for? Shouldn't California state have the ability to make things happen for its own rail projects?

You'd feel differently if it was YOUR land being seized.

I'm sure I would, and I'm not blaming the land owners for being angry or litigious. But it's the government's job to decide where the line is drawn and when your options run out.

And whose taxes are being seized?

Seized but you are paid market value for it. That’s fair.

I'm left wondering: how much would a hyperloop cost to span the same distance? At 2 or 3 times the speed?

They've already started it? Never too late to stop a disaster. It'll be $150B by the time it's finished.

Is there hyperloop? No.

Reply on news.ycombinator.com