Cost overruns in the oil sands

  As part of writing a book about the Panama Canal, Carlos and I had to do a lot of research into cost overruns on big infrastructure projects.  We found the usual horror stories.  The Panama Canal went over-budget by a factor of 2.0.  BART exceeded projections by 1.6, and the Washington Metro hit 1.8.  Miami’s Metrorail jumped the limit by 2.1, the Big Dig ran over by 2.9, and the Brooklyn-Queen Expressway beat its budget by an astonishing 5.1. 

Sounds bad, right?  And in the case of the BQE, it is bad.  But you haven't seen overruns until you've seen the private sector try build multi-billion dollar stuff in the Albertan oil sands.  Suncor's Millenium project ran over by 1.6, about the same as BART.  Shell's expansion of the Muskeg River mine has hit 2.2, and counting.  In 2008, Syncrude's Stage 3 expansion passed 2.4.  The Canadian Association of Petroleum Producers estimates that capital costs for ongoing projects are going to average as much as 2.7 times the 2005 projections. 

Not quite the Big Dig, but making the Panama Canal look good.  I have some thoughts as to why this happens, but before I reveal them, what do you think?

Comments

[Bernard Guerrero]

Crappy environment? I'm going to guess that there's a fair amount of technically challenging work that's being done in what amounts to the middle of nowhere. The cost associated with getting the required talent and equipment in place is the sort of thing that might reasonably be underestimated, at least the first few times.

Idiosyncratic issues with each deveopment? (Or, crappy scale?) These tend to be relatively small projects still, right? A lot of development work can be contracted out to experienced specialist firms in other areas of oil drilling. There are loads of specialist deep sea drilling, shallow sea drilling, supply and logitistics companies operating in, say, the Gulf. Not only are the technical issues better understood, but the scale exists for specialists to operate profitably within the niches those technical issues create. I suspect everything being done with regards to the oil sands is de novo and custom built.

[Dennis Brennan]

Ordinarily I would have said unexpected opportunistic rent-seeking by the locals... but there aren't any locals out there, are there?

[Charlie Stross]

For real overruns, there's always the Scottish Parliament at Holyrood to beat most comers -- originally costed at between £10- and £40M, the final bill came in at £414M (three years late on a two year project, to add insult to injury).

However, that says nothing about Alberta.

Did the oil companies want to run up a big operating loss to write off against profits?

[Bernard Guerrero]

The companies invloved tend to be JVs of bigger players, so I think there'd be a coordination issue as far as purposefully running up losses.

I'd also guess that the reasons for cost overruns in government projects in well-developed areas like, say, Boston tend to be different from what we're seeing in Alberta.

[Randy McDonald]

I know for a fact that labour supply has been a huge issue, not only in the oil fields but in Alberta as a whole. Workers at Tim Horton's coffee shops get paid $C15 an hour and scholarships to keep them from moving north, while migrants have come from all over central and eastern Canada. Wages are terrifically high, such that professionals can drop everything, go west, and come back richer than they would have otherwise.

[Bernard Guerrero]

How much richer?

[Randy McDonald]

From the anecdotal stories I've heard, tens of thousands of dollars richer.

The living standards are atrocious, of course, a dozen or so people crowded into a bungalow, but still, it's the standard tradeoff made by migrant workers everywhere. Atlantic Canadians are used to this kind of thing, but it's very unusual for Ontarians, whose province was usually the _destination_ for this class of people.