Doing a bit of research on the authors of the white roofs article I referenced a few days ago, I found a very interesting essay by Richard Muller in a 2002 issue of Technology Review. In it, he cites an essay by Arthur Rosenfeld (along with TM Kaarsberg and JJ Romm) entitled "Efficiency of Energy Use," (unfortunately not available online, as far as I could find) in the Macmillan Encyclopedia of Energy in which the authors argue that:
From 1845 to the present, the amount of energy required to produce the same amount of gross national product has steadily decreased at the rate of about 1 percent per year. This is not quite as spectacular as Moores Law of integrated circuits, but it has been tested over a longer period of time. One percent per year yields a factor of 2.7 when compounded over 100 years. It took 56 BTUs (59,000 joules) of energy consumption to produce one (1992) dollar of GNP in 1845. By 1998, the same dollar required only 12.5 BTUs (13,200 joules).
The one percent/year improvement fluctuates a bit, but has remained largely consistent except for during the early 1970s, when it jumped to 4 percent per year. Rosenfeld argues that, with a little bit of effort and government encouragement, we could easily sustain a 2 percent improvement in efficiency. This difference may sound small, but has staggering results when looked at over time.
With 1 percent annual improvement, population stablizing at around 10 billion, and overall increase in standards of living to US/EU levels, the globe would be using 40 percent more energy in 2100 than today. But by bumping up overall efficiency improvement to 2 percent averaged over the next century, Muller calculates that we'd actually end up using half our current levels of energy.
The good news is that, like Moore's Law, there is an observable, consistent improvement in energy efficiency over time. The better news is that this has happened largely without making a focused effort. Imagine if there was competition regarding efficiency as aggressive as that for processor improvements...
If our asses are going to be saves, this is how it's going to be done. I'm a technological determinist, I am!
I think this is possibly the single most important post on WorldChanging in some time. Energy, to me, seems like the most important of the sustainability issues to address, because it's the one likely to begin having the most devastating consequences, first. But it also tends to be the one that makes people throw up their hands the fastest. Analysis like this is absolutely crucial to focusing our efforts and generating the political will to solve this problem (currently at an all-time low).
I agree, Jesse. I'm now trying to get an interview with Arthur Rosenfeld, and a copy of a new version of this paper!