Cost of public works projects

On Friday I was walking by the Alaskan Way Viaduct thinking about the cost of the project and it struck me: given the monumental expense of these projects how the heck did they ever happen in the 1st place?

The Alaskan Way Tunnel project is estimated to cost 4.2 billion dollars. For comparison the Hoover Damn cost 49M in 1931 dollars (~ 700M in 2011 dollars, adjusted via the CPI). Yes, I can hear the argument these are not identical projects. No kidding. But to argue that the Alaskan Way Tunnel is of greater scope seems incredulous. Is this yet another example of Baumol’s cost disease? I would have thought that advances in technology would have made construction cheaper (as opposed to teaching).

Some theories that pop to time are that material costs are much higher in 2011 than they were way back when. Or maybe it is the cost of labor is much higher now than it was back then? I can’t believe it is graft since the numbers are simply too large.

My guess is that building anything in a dense urban area is an order of magnitude more expensive.

Not a fan of these huge projects designed to alleviate auto congestion, basically amount to automobiles subsidies…

Not a fan of these huge projects designed to alleviate auto congestion, basically amount to automobiles subsidies…

Unfortunately similar arguments tend to be raised against anything that alleviates congestion by adding public transit.

Even if you don’t subsidize public transit, just not subsidizing automobiles would help. Cities building parking garages on absolutely prime real estate and then charging low rates drives me nuts too.

Inflation adjusts for same-good value, but there’s a bunch of other stuff that leads to confusion on big-ticket items like dams.

For example, someone buying a car then vs. a car now is straightforward - the car is way better and a smaller fraction of income. To compare “scope” of a big public project, though, I think it’s more useful to look at “how much of society’s total output did that take?” Remember dollars are just claims on real, physical production.

This has 1931 real-dollar GDP at ~830 billion. The Hoover Dam at 700 million was nearly a full percentage point of GDP - 0.85%.
2011 GDP is 14 trillion, so the 4 billion tunnel is 0.03% of GDP - 1/33rd of a percentage point. It’s way, way cheaper in terms of what society has to give up to build it.

Now, back to straight dollar comparisions:

  1. The technology used to construct the tunnel is way, way better than what they’re using to build the dam. The dam is basically a gigantic amount of labor and concrete; your chief costs are unskilled labor and dirt, back when there was a ton of cheap unskilled labor. The tunnel, by contrast, is going to use some crazy-ass boring high-technology machine; it’s not going to be particularly labor intensive. I don’t even know if it’d be possible back in 1931 given all the geographical constraints.
  2. The dam is, well, a dam. You build the scaffolding, poor the concrete, and you’re done. There was little concern for environmental impact. There’s no population around, and you can basically do whatever you want to the landscape. By contrast, the tunnel has to be drilled under an existing downtown area under century-old buildings, without knocking down anything, and has to be built to stand up to a 9.0 earthquake that liquifies the soil.

I think the project is a ridiculous boondoggle, because it’s not even projected to reduce congestion, but it’s incredibly ambitious for something so pointless.

The Alaska Way tunnel project (in Seattle FWIW) is actually replacing a two deck freeway that was declared dangerous way back in 2001. The big issue to me is not how much its going to cost, but that it took 10 years for people to decide how to do it. Ultimately they chose a tunnel over a replacement surface or raised freeway. Of course a tunnel is going to be more expensive.

But in general Seattle area public works projects are very expensive. When I think of the billions sunk into light rail (which I voted for) compared to what little we actually have 15 years later, I get somewhat ill.

Public works projects in any area that is already built up are tremendously expensive.

The light rail (and the South Lake Union Streetcar) are absolutely mind blowingly expensive for what you get, but I think the light rail (and streetcar) are helping with the concept of downtown-as-pedestrian-friendly. I have lots of friends that can afford to use their cars but opt to take buses or other light rail.

For me it’s a 10 minute ‘walk’ to the airport, as long as I don’t have to catch some crazy flight (e.g. tomorrow I have a 6am flight to Montreal which means there’s no way I can take the light rail since it doesn’t run at those hours).

I think the bigger issue is enforcement and cost, in particular the streetcar is ridiculously expensive ($2?) for a very short ride. There have been cases where I’ve beaten the street car while walking.

If the streetcar were ubiquitous, i.e. went all the way up to the U-distrct and also into cap hill then it would be a lot more interesting.

That said, ridership on both has shot up considerably, and the light rail is absolutely packed after big games.

Not to address the rest of your post, which is interesting, but this is off by close to a factor of 10 - 1% of 830 billion is 8.3 billion.

One of the problems they’re gonna have with the tunnel is going to be the fact that they’re gonna need to dig down pretty deep to get the thing settled in solid ground.

That’s actually one of the major engineering mistakes of the Viaduct, in that they basically built the thing on top of loose fill, rather than bedrock, which makes the whole thing unstable.

Jason makes some decent points about the technological feat of constructing a giant dam vs. the tunnel, although I think he is perhaps underestimating the technology required to construct the hoover dam. The scale of that construction required some technological innovation.

That being said, if you want a real “Holy shit, that crap is impossible” engineering feat from the good ol’ days, you should look at the Chesapeake bay bridge-tunnel. Constructed in 1964, it was (and continues to be) one of the most impressive engineering feats of the human race. I think most folks don’t really realize this, because, hey… bridges and tunnel, big whoop. But constructing it across the mouth of a bay, with constant, strong, shifting currents made its construction something that many people doubted could be done at all.

They built it in 4 years, working through hurricanes, noreasters, and all kinds of other crazy crap. I’m not sure how much it actually cost them to build it though.

According to wikipedia it cost $200m in bonds issued in 1960. There was no public money involved - the bonds were paid off entirely by tolls, as have all expenses and expansions to it since. In 2005 there was talk of replacing the tunnel portion with a larger version, with an estimated cost of $900m.

And note that the population in 1931 was between one half and one third of the population in 2011 and the CPI adjusted numbers are almost spot on (begging the question: what-the-heck!).

BTW: does anyone have a good link on the estimated cost of the 520 expansion project?

The Big Dig was as I understand it (no reference at hand) the most costly public works project in human history. It got that way through massive and thoroughly pervasive corruption at all levels, from the state house to the prime contractor to the rats in the tunnels.

Graft? Really? I can’t believe that accounts for the monumental differences.

The pyramids probably weren’t all that straightforward.

Whoops, you’re right, Hoover Dam was 0.085% of GDP, tunnel is 0.28% of GDP. Assuming no cost overruns - ha, I’m so funny. That thing is going to be murder by the time it’s done.

Undoubtedly there was a great deal of sheer incompetence involved in the Big Dig, but it’s also very clear that there was a great deal of corruption and a variety of other criminal activities involved as well, including simple fraud.

This may be an incredibly naive question, but why aren’t these monumental construction projects bid on a fixed price basis? i.e. whoever the contractor is that wins the bid actually had to deliver at that price, with no over-runs, etc.?

Probably because they need to not fall down or explode unexpectedly.

California High Speed Rail…

You pass a bond for one price and within a few years the price has exploded.

The California High-Speed Rail Authority’s new cost estimates released Tuesday show the initial stretch of construction between Merced and Bakersfield will cost $10 billion to $13.9 billion depending on how it’s built. Project planners had previously pegged the section at $6.8 billion.

If the cost of the entire project balloons at the same pace as the Central Valley section, the San Francisco-to-Anaheim railroad would cost from $63 billion to $87 billion, similar to what independent analysts have been predicting. And those figures do not include inflation, which could push the final cost toward a staggering $100 billion.

When California voters approved the project in 2008, the state said it would cost $33 billion, but it soared to $43 billion a year later – already making it the single largest public works project in the nation. Even now, the state only has about one-fourth of the money needed to fund the entire rail line and no clear plan on how to secure the rest.
“It leaves what looks like a potential tremendous burden on the California taxpayer,” said Assemblyman Jerry Hill, D-San Mateo, whose Peninsula district has led the chorus of boos for the project. “We can’t double our debt obligations for the high-speed rail system. That would all be done at the expense of education, and health and human services, things that people (need).”