Alerted by Jonah Goldberg, I read Patrick Deneen’s post about the “peak oil and global warming culture wars”. My reaction to the original post was basically that people losing sleep over peak oil should switch to decaf. I said:
Crude oil production will reach a maximum at some point in the future. I don’t know when that will happen, and the record of those who have tried to forecast this has not been very good over the past 70 years or so. When that happens, the price will probably rise. We will develop technological alternatives and find substitute fuels. It’s not time to start burying Krugerrands in the backyard.
Mori Dinauer at The American Prospect said:
There’s plenty to mock in this … As you probably know, in 1956 a geologist with Shell Oil named M. King Hubbert made a forecast that oil production in the United States would peak sometime in the late 1960s or early 70s. Oil production in the United States did peak in 1970, so it isn’t really defensible to say that nobody knows when world oil production will peak. …
Hubbert did make this call. Dinauer fails to note, however, that in 1974 Hubbert also predicted that global oil production would peak in about 1995. Whoops. (There are good reasons why it’s more feasible to predict peak production in a very well-understood geography, such as the US, than for the world as a whole.)
You will not be shocked to learn that, in fact, lots of forecasts for peak oil have been made over the years. The U.S. Department of Energy (DOE) identified 36 such forecasts published between 1972 and 2004. I conducted a quick analysis of the forecast for peak oil over time (1):
Notice any pattern in this data? Roughly speaking, forecasts indicate that we are 20 – 30 years from peak oil today, just as forecasts generally indicated that we were 20 – 30 years from peak oil throughout the 1970s and 80s.
Unsurprisingly, the DOE has taken a serious look at this question. Their best guess (and they are rigorous enough to put a range of many decades on this) is that peak production will be reached sometime in the middle of this century. The International Energy Agency projects that production will continue to increase at least through 2030. So does OPEC.
Oh, and since oil now costs over $120 a barrel, does that mean oil has peaked, Jim?
On an inflation-adjusted basis, this is almost exactly what oil cost in 1980. Does that mean oil production peaked in 1980?
I’ll also note that while the world spent about 6 percent of its total economic output on oil in 1980, this is down to about 3.5 percent today. Maybe this is why I see no observable signs of the collapse of modern civilization resulting from the current run-up in oil prices.
Clark Stooksbury at The American Conservative had this to say:
So we will know when oil has peaked because prices will rise. Too bad he didn’t tell President Bush –it could have spared us the ugly sight of seeing the leader of the free world ‘debasing himself by begging our Saudi Masters to turn up the pumps.
I said that when oil-production peaks, the price will probably rise. I didn’t say there are no other conditions under which prices will rise (see prior comments about 1980). So, no, we won’t “know when oil has peaked because prices will rise.” I also seem to recall other American presidents “begging our Saudi Masters to turn up the pumps” between the years 1973 and 1980, when oil production had not peaked. Having a supply chain that originates in unstable desert kingdoms is not the same thing as reaching peak oil.
James Poulos at Postmodern Conservative makes the sensible observation that:
If I had to make a guess, sovereign state control over oil resources — call it ‘oil nationalism’ — will be a much bigger factor going forward whether we’re under peak oil conditions or not
Whether or not this increases as compared to today, sovereign control over oil resources is certainly a huge issue for the world.
1: Average of forecasts in any year for which there is one or more forecast. Excludes Ivanhoe 1994, since this only makes a projection for OPEC production. Uses the maximum point estimate for any study (2060) as the forecast for any study that projected “No visible peak” or “Beyond year X.”