5 ms·

People living in the East Bay (SF Bay area) might be interested to know that the 4th bore of the Caldecott Tunnel (completed 2013) was the largest recipient of

People living in the East Bay (SF Bay area) might be interested to know that the 4th bore of the Caldecott Tunnel (completed 2013) was the largest recipient of Federal stimulus funds for infrastructure in the US ([1] via [2]). My own recollection is that the project was nearing the final planning stages in 2008, but the recession wiped out much of the state and local funding that had been earmarked for the project. The project was about to be cancelled, but Federal stimulus funding came to the rescue. [1] http://www.sfgate.com/bayarea/article/Work-begins-on-Caldecott-Tunnel-s-4th-bore-3275155.php http://www.sfgate.com/bayarea/article/Work-begins-on-Caldeco... [2] https://en.wikipedia.org/wiki/Caldecott_Tunnel#Fourth_bore https://en.wikipedia.org/wiki/Caldecott_Tunnel#Fourth_bore (Edit: in an earlier version, I claimed that the project was "shovel-ready" in 2008, but since construction didn't start until 2010, I was obviously not correct).

m_mueller 10y ago
Is my interpretation correct that the main problem in US public infrastructure is that politicians can just move around the funding of already running projects? Or is it bad incentives when doing estimates such that it can't be planned properly? I'm asking because in Switzerland we have just completed two huge infrastructure projects, one for 10B and one for 2B, with hardly any budget problems. Cost of 1km of rail tunnel is at a reasonable 140M (for two ways, one bore each). So we know it's possible, I'm just wondering what's the difference.
- CalChris 10y ago
  That's just ridiculously less expensive. Railway bores are gonna be larger (6.2m vs 4.7m) and then you have two bores. They're about the same length too. Bore 4 of the Caldecott is 1033 m.
  - m_mueller 10y ago
    Yep. It's not exactly easy task either. I'm talking about [1], the new largest train tunnel at the base of the alps, a very difficult terrain to tunnel through, and [2], a new train line, including a new underground train station, right through the fully developed center of Zurich. In a country with among the highest salaries in the world. I'm just still trying to wrap my head around what the hell the US is doing with infrastructure projects. Too many cooks? Constantly changing administration? Corrupt building industry and regulatory bodies? It doesn't seem just a federal problem either, at least in the big US cities there's news left and right about billions wasted in what I'd call small to moderate extensions of public transit infrastructure. [1] https://en.wikipedia.org/wiki/Gotthard_Base_Tunnel https://en.wikipedia.org/wiki/Gotthard_Base_Tunnel [2] https://en.wikipedia.org/wiki/Weinberg_Tunnel https://en.wikipedia.org/wiki/Weinberg_Tunnel
    - sandworm101 10y ago
      >> Too many cooks? Constantly changing administration? Corrupt building industry and regulatory bodies? I'd say not too many cooks, but too many head chefs. From my contact with federal projects there always seems to be too many bosses. Dozens of people/groups think they are the final word on the project. None work with each other. Instead of all sitting in a room to hash things out, project plans are emailed/mailed around. Each then sits on the project for weeks, or months if they feel they need public participation/input. By the time each stakeholder agency does their thing a year has past. Any change then restarts the email chain. European governments seem better at agreeing general concepts and then appointing educated individuals to act as reps. Meetings then move far more quickly.
      - m_mueller 10y ago
        I'm guessing these administrations are the result of laws and/or regulations, right? And the US probably has way more regulation, simply because there is a bigger pot to pay politicians and lobbyists from, and they all need something to do/to promise their voters? Here's an idea: What if every new piece of law and regulation came with a mandatory expiration date - say 10 years. After that date it must pass the legislator again, at which point a new date is set based on an exponential backoff algorithm. At each new vote I'd expect there to be some evaluation of the effects of that piece of legislation. Now obviously, you still have the problem of strictly partisan politicians, but if you could even just have 5% of legislators not being corrupted by that it would be enough to swing the majority to one side or the other, based on facts.
        sandworm101 10y ago
        It's not the amount of regulation on the books. Europe has more. It's a cultural thing. Americans don't trust their governments in the same way Europeans do. Even the most liberal-minded democrats still want government to be constantly checked. That makes those laws that are there very difficult to work with. Everyone sees is as his or her duty to watch what the government is doing. I don't mean just citizens. The American system of divided government often sees state lawmakers at odds with their federal government, doubly so when the project cross state lines. Everyone on the project tiptoes around slowly thinking any misstep will result in legal troubles. Europeans are far more trusting of democracy, taking such matters to the ballot box rather than the lawyer's office. An example: There is a regulation stating that federal projects must notify and consult affected native tribes. Fine. That's sounds reasonable. But in practice "affected tribes" has become all tribes. Rather than risk getting it wrong, some feel they must officially inform and consult every tribe in the nation. There is a few weeks delay. Here is the handbook for that reg. Note that this is not the regulation but a guideline on how to obey it. The reg itself is very simple (It's actually a body of several regs, but they all do basically the same thing). What makes it complicated is all the stuff over and above the wording, the stuff you do more out of fear than legal requirement. That stuff gets published in a *.gov PDF and the next thing you know everyone thinks they have to obey every letter of the PDF. The PDF is now a defacto reg and the cycle begins again. http://www.achp.gov/pdfs/consultation-with-indian-tribes-handbook-june-2012.pdf http://www.achp.gov/pdfs/consultation-with-indian-tribes-han...
- mturmon 10y ago
  I don't think you can compare projects at this level of detail. My understanding of the Swiss tunnels (congratulations!) is that they were through hard rock which provides predictable support for the tunnel. Tunnels in softer or less consistent materials have much less predictable costs, as do projects involving water.
  - m_mueller 10y ago
    Tunneling through the alps isn't exactly easy, it's actually harder than through a normal sediment. Well, maybe cost wise a weak sediment may be more expensive (but not by a factor of 20x, more like 1.5-2x), but vertical rock formations bring all kinds of surprises, which leads to accidents including deaths, which leads to insurance premiums. To illustrate, here's a normal sediment: ----------------- 20y ----------------- 100y ---------------- .... here are the alps: | 10M years | 10.1M years | 12M years | .... Edit: You can also compare to the other link I posted in a sister comment, about a new tunnel through one of Zurich's hills. That project was actually less expensive per km than the Gotthard base tunnel, so I don't think the rock is the problem.

m_mueller 10y ago

Is my interpretation correct that the main problem in US public infrastructure is that politicians can just move around the funding of already running projects? Or is it bad incentives when doing estimates such that it can't be planned properly? I'm asking because in Switzerland we have just completed two huge infrastructure projects, one for 10B and one for 2B, with hardly any budget problems. Cost of 1km of rail tunnel is at a reasonable 140M (for two ways, one bore each). So we know it's possible, I'm just wondering what's the difference.

CalChris 10y ago
That's just ridiculously less expensive. Railway bores are gonna be larger (6.2m vs 4.7m) and then you have two bores. They're about the same length too. Bore 4 of the Caldecott is 1033 m.
- m_mueller 10y ago
  Yep. It's not exactly easy task either. I'm talking about [1], the new largest train tunnel at the base of the alps, a very difficult terrain to tunnel through, and [2], a new train line, including a new underground train station, right through the fully developed center of Zurich. In a country with among the highest salaries in the world. I'm just still trying to wrap my head around what the hell the US is doing with infrastructure projects. Too many cooks? Constantly changing administration? Corrupt building industry and regulatory bodies? It doesn't seem just a federal problem either, at least in the big US cities there's news left and right about billions wasted in what I'd call small to moderate extensions of public transit infrastructure. [1] https://en.wikipedia.org/wiki/Gotthard_Base_Tunnel https://en.wikipedia.org/wiki/Gotthard_Base_Tunnel [2] https://en.wikipedia.org/wiki/Weinberg_Tunnel https://en.wikipedia.org/wiki/Weinberg_Tunnel
  - sandworm101 10y ago
    >> Too many cooks? Constantly changing administration? Corrupt building industry and regulatory bodies? I'd say not too many cooks, but too many head chefs. From my contact with federal projects there always seems to be too many bosses. Dozens of people/groups think they are the final word on the project. None work with each other. Instead of all sitting in a room to hash things out, project plans are emailed/mailed around. Each then sits on the project for weeks, or months if they feel they need public participation/input. By the time each stakeholder agency does their thing a year has past. Any change then restarts the email chain. European governments seem better at agreeing general concepts and then appointing educated individuals to act as reps. Meetings then move far more quickly.
    - m_mueller 10y ago
      I'm guessing these administrations are the result of laws and/or regulations, right? And the US probably has way more regulation, simply because there is a bigger pot to pay politicians and lobbyists from, and they all need something to do/to promise their voters? Here's an idea: What if every new piece of law and regulation came with a mandatory expiration date - say 10 years. After that date it must pass the legislator again, at which point a new date is set based on an exponential backoff algorithm. At each new vote I'd expect there to be some evaluation of the effects of that piece of legislation. Now obviously, you still have the problem of strictly partisan politicians, but if you could even just have 5% of legislators not being corrupted by that it would be enough to swing the majority to one side or the other, based on facts.
      - sandworm101 10y ago
        It's not the amount of regulation on the books. Europe has more. It's a cultural thing. Americans don't trust their governments in the same way Europeans do. Even the most liberal-minded democrats still want government to be constantly checked. That makes those laws that are there very difficult to work with. Everyone sees is as his or her duty to watch what the government is doing. I don't mean just citizens. The American system of divided government often sees state lawmakers at odds with their federal government, doubly so when the project cross state lines. Everyone on the project tiptoes around slowly thinking any misstep will result in legal troubles. Europeans are far more trusting of democracy, taking such matters to the ballot box rather than the lawyer's office. An example: There is a regulation stating that federal projects must notify and consult affected native tribes. Fine. That's sounds reasonable. But in practice "affected tribes" has become all tribes. Rather than risk getting it wrong, some feel they must officially inform and consult every tribe in the nation. There is a few weeks delay. Here is the handbook for that reg. Note that this is not the regulation but a guideline on how to obey it. The reg itself is very simple (It's actually a body of several regs, but they all do basically the same thing). What makes it complicated is all the stuff over and above the wording, the stuff you do more out of fear than legal requirement. That stuff gets published in a *.gov PDF and the next thing you know everyone thinks they have to obey every letter of the PDF. The PDF is now a defacto reg and the cycle begins again. http://www.achp.gov/pdfs/consultation-with-indian-tribes-handbook-june-2012.pdf http://www.achp.gov/pdfs/consultation-with-indian-tribes-han...
mturmon 10y ago
I don't think you can compare projects at this level of detail. My understanding of the Swiss tunnels (congratulations!) is that they were through hard rock which provides predictable support for the tunnel. Tunnels in softer or less consistent materials have much less predictable costs, as do projects involving water.
- m_mueller 10y ago
  Tunneling through the alps isn't exactly easy, it's actually harder than through a normal sediment. Well, maybe cost wise a weak sediment may be more expensive (but not by a factor of 20x, more like 1.5-2x), but vertical rock formations bring all kinds of surprises, which leads to accidents including deaths, which leads to insurance premiums. To illustrate, here's a normal sediment: ----------------- 20y ----------------- 100y ---------------- .... here are the alps: | 10M years | 10.1M years | 12M years | .... Edit: You can also compare to the other link I posted in a sister comment, about a new tunnel through one of Zurich's hills. That project was actually less expensive per km than the Gotthard base tunnel, so I don't think the rock is the problem.