Hacker News Reader

>... I also don't agree that a project should be killed because it's cost has increased ~$20B over what was approved in 2008.

You aren't even close. The fiscal impact statement of what was voted on in 2008 was:

>...State costs of about $19.4 billion, assuming 30 years to pay off both principal ($9.95 billion) and interest ($9.5 billion) costs of the bonds. Payments of about $647 million per year. When constructed, additional unknown costs, probably in excess of $1 billion a year, to operate and maintain a high-speed train system. The costs would be at least partially, and potentially fully, offset by passenger fare revenues, depending on ridership.[6]

>...First, 10 years of inflation alone from 2008 to 2019 accounts for $10B of price increases.

How are you calculating that? The cumulative rate of inflation from 2008 is < 17%.

>...The annual payments on the bond would be around $1B/year, out of a total annual budget of $200B

How are you calculating the bond payment? (It sounds very low.) In the original ballot measure, where 10 billion was issued in bonds, the payments on the 30 year bonds was estimated to be about $647 million per year. Your total annual budget number is misleading - I've read that only about 22% of the CA budget is discretionary.

>...the laughable assertion that a train from Bakersfield to Merced is a good idea is clearly a face saving measure.

I agree. This is now a train to nowhere and makes people leery about any spending on infrastructure.

> You aren't even close. The fiscal impact statement of what was voted on in 2008 was: ... $10B

The $10B was not billed as the total cost of the project, just bond that voters approved. The rest of the project was to be funded with "TBD" sources. I believe the initial cost estimates were in the $40B range. [1]

> How are you calculating that? The cumulative rate of inflation from 2008 is < 17%.

I assumed 2% annual inflation over 10 years, inflating an initial cost estimate of $50B.

[1] https://www.sfgate.com/opinion/article/Yes-on-Prop-1A-A-21st...: "The project is expected to cost $40 billion and take eight years to complete, according to proponents."

>...The rest of the project was to be funded with "TBD" sources. I believe the initial cost estimates were in the $40B range.

Thanks for pointing that out, I didn't realize that. I wonder how many of the people who voted for the bonds knew that there was an estimated extra 30 billion in cost.

Even going from an estimated 40 billion to 77 billion is still a huge increase. The original article mentioned that the state estimated the cost might actually be $98.1 billion.

I don't have a good sense for how the typical voter contexualizes these things. The numbers are so big and abstract I don't see how you can tell $20B from $60B as a voter.

I think though, if people had known in 2008 that not even 100 miles would be completed by 2019, that might have swayed some minds. In my view, the big tragedy here is the painfully slow construction, not so much the cost.

> California has 40 million residents, so that comes out to a $2000 loan on a per person basis.

40% of Californians don't pay any state income tax so it's closer to $5000 per person. As for the project reducing unemployment- good luck with that being they are only going to finish a 110 mile section between Merced and Bakersfield.

$77 billion doesn’t sound like a big number because of the absolute amount of money involved. It sounds like a big number because similar projects in other places cost far, far less. That number represents an incredible amount of waste. I think they should focus on getting the project to cost a reasonable amount and then do it, rather than either accepting it as-is or abandoning it.

How can these projects/plans/programs get the costs down?

Beats me, but it’s clearly possible.

Why? Maybe it's just what this costs. Better than dreaming up a lower number that has no connection to reality. (Just as software projects are estimated based on wishes instead of hard data.)

As trite is to use Musk as an example, BoringCo at least tries to do something to bring costs down. (And they put some money toward that goal.) But a completely powerless and political project planning committee can only make more and more sheets of wasted paper, I'm afraid.

Someone mentioned how highways are also very expensive. I don't know why. ( https://medium.com/@TimSylvester/i-agree-it-sounds-astronomi... ) But without changing the fundamentals, without innovation the costs won't come down.

We can probably do a very minimal estimate by simply using average/median income, and estimating how much worker hours we need to plan it, review it, procure/tender it, get stuff there (from surveyor equipment and machines to people going there), do the earthworks, fill in the rocks and gravel that will serve as the foundation, then pour the concrete (or better yet use prefab/precast), then surface works, then put there all the signs, safety stuff (railings), environmental protection stuff (noise barrier), and maybe that's what it costs.

So, as long as we can't streamline, uniformize and scale up these processes, it'll be just very labor intensive and expensive.

"...because similar projects in other places cost far, far less."

If Alice is doing something and Bob says it's impossible, I don't need to know how to do it to point out that Bob is wrong and it's actually possible.

(And it's not just a matter of saving costs through environmental destruction and labor abuse. Infrastructure costs much less in other first-world countries too.)

But other places have other price tags.

Sure, there are problems - for example there is a frequently referenced study by one EU oversight institution from 2013. And it turns out that highways built between 2000 and 2012 sponsored by the EU were cheapest in Germany, where components were more expensive, than in Spain (where everything from labor to land and concrete were cheaper, including compliance to environmental regulations).

Of course that's corruption. But the US has probably a different problem. Cost inflation, because the market bears it. Supply and demand. States pay a lot, so suppliers ask a lot.

To put it into perspective, the Hoover Dam cost $49M, which is $750M in today's dollars [1]. They claim these train tracks cost over 100X more.

One estimate of the cost to build the trans-continental railroad, including the Central Pacific ($47M) and Union Pacific ($77.5M) [2] is $3.5B in today's dollars, so this railroad costs 22X what two cross-country lines once cost.

1: https://www.marketplace.org/2010/05/28/sustainability/why-we...

2: http://cprr.org/Museum/FAQs.html#Cost

Hey, if you can:

-collapse the global economy so that incredibly skilled engineers and tradespeople are desperate to work for little more than what it takes to feed their families and live in on-site housing

-reduce safety regulations enough that it's acceptable to have people regularly dying during construction

-build the train out in the middle of the desert on government-owned land instead of running it through major cities

Then yeah, you could make these trains pretty cheap.

Crossrail[1] is currently projected to cost £15.4bn (~$19.8bn) for 73 miles of non-high-speed rail (~90mph).

For the 520 miles California was proposing, you'd be looking at £110bn (~$140bn) and that's ignoring the fact that London has no mountain ranges to contend with and that HSR requires much stricter paths.

In that context, $77bn looks like an absolute bargain.

[1] https://en.wikipedia.org/wiki/Crossrail

I’m not sure an urban line with substantial underground sections and a bunch of new stations is entirely comparable.

The Jinghu HSR rail line from Beijing to Shanghai spans 1318 km and cost 32 billion USD. It operates at an average speed of 180 MPH. The California HSR plan is considerably shorter, considerably slower, and would cost over twice as much.

It would probably have a fraction of the casualties and environmental impacts in construction though. China has different standards to the modern world.

Well, the difference is that China has useful high speed rail and California does not.

That's why California has reason to waster few billions then abandon the whole project. Great California republic!

That’s good and well, but remember that China has much lower labor costs, and fewer environmental protections (CEQA). I don’t think the low price of Chinese rail should deter HSR in California.

Also, we literally __voted__ for the project.