Note: This was written in 2007. I like to keep my old predictions around. It tends to keep me humble.

A friend of mine who is involved in the oil industry recently pointed out something I consider important in understanding the future of oil. He said that we are running out of cheap oil rather than running out of oil. There is still a lot of oil out there, but it's harder and more expensive to get at than the oil we've extracted so far. That means that while oil prices will undoubtedly bounce around, in the long term they are going to trend up. That's pretty much inevitable barring some kind of breakthrough in extraction technology, but it isn't necessarily a bad thing. High and unstable oil prices may help wean us off of oil gradually, so that the transition doesn't leave large parts of industrial society starving in the dark.

First, we are running out of cheap oil. We used the cheap oil that was easy to get to first. That only makes sense. Now we're increasingly using the oil that it didn't make sense to use when cheaper sources were available. For example, it didn't make sense to extract oil from the tar sands up in Alberta until the price of a barrel of oil got to around twenty dollars per barrel. Now it makes sense to extract that oil, and oil companies are doing so in a big way.

High oil prices have an odd impact on demand for oil. Remember the bit about supply and demand you learned back in school? According to that, if prices rise, supply goes up and demand falls. Well, it sort of works that way with oil, but with some complications. In the short term, high oil prices don't have much impact on demand or supply. Think about how you would react if gas prices went up to $5 per gallon. Well, if you are anything like me you would drive as little as you could, but you've got to get to work, and you have to get groceries. There is a limit to what you can do to reduce your demand in the short term.

The same is true with supply of oil. It takes years and billions of dollars to expand supply. Getting that oil out of the Alberta tar sands takes a lot of investment before the first barrel of oil flows.

So, is this supply and demand thing a bunch of baloney when it comes to oil? No, not in the long term. High prices have a small immediate impact on demand, but if they stay high long enough they have much larger long-term impacts. If prices go high, people initially drive less and fly less. If those prices stay high long enough, people buy more fuel-efficient cars. If average miles per gallon goes from 20 to 35, that has a huge impact on demand for oil. In the even longer term, with higher prices eventually people find that high commuting costs more than offset lower housing costs in the suburbs. Urban sprawl slows down and reverses if prices stay high long enough. Industrial processes get reworked to use less energy. New houses get a lot more insulation.

Prices aren't high enough yet to trigger major overhauls in energy efficiency, but they'll probably get there.

Oil prices do respond to supply and demand, but they take a while to do it. If oil prices stay high long enough, people make capital investments to off-set those costs: buying a smaller car, buying a smaller and better insulated house closer to work. Companies and governments also invest in increased supply-reopening wells that weren't worth operating at the lower costs, investing in tar sands, building solar cell manufacturing plants. All of that means that if oil prices stay high long enough that people decide the higher prices are here stay, demand goes lower and supply of alternatives goes higher.

Normally, high oil prices would be self-correcting, and oil prices would eventually drop below where they would have been without the price spike. That's what happened in the early 80's. Oil prices peaked at $35 a barrel in 1981 and then dropped below $10 in 1986 before starting back up. As the example of the early 80's shows, prices fall to a point lower than they would have been without the price spike. Not only that, but they stay depressed relative to what they would have been without the price spike for ten to twenty years or longer. Once people have invested in a fuel-efficient car or house, they don't throw those things away if energy prices go down. They don't usually tear out added insulation.

If a company has already invested in oil wells or alternate energy sources they don't necessarily shut those investments down if oil prices go down to the point where they aren't going to get all of their investment back. The prices have to go down to where they are not getting any of their investment back and probably down to where they are actually losing money on day-to-day operations.

In the long-term, demand for oil should fall and so should prices. Nothing to worry about in the long term, right? Normally that would be true, but this time it may not be. China and India are industrializing quickly, and unfortunately their growing middle classes seem to equate wealth with the very inefficient Western style of using energy. In order for overall demand to shrink, demand in already developed countries would have to more than offset the increased demand in China and India. If two billion Chinese and Indians start to use natural resources, including oil at the rate the US, Japan, and Western Europe use it, then it gets very difficult to see supply and demand coming into balance without some wrenching changes.

Fortunately, we do have some options that may help out. We are nowhere close to as efficient in using oil as we will be forced to be if oil prices go high enough. We tightened up in the 70s and early 80s, but then got sloppy again when oil prices collapsed. Just buying smaller, more fuel efficient cars would help a lot. As hybrids get less expensive they'll help too. A new generation of 'plug-in hybrids may have even more impact. The idea is that you charge up a battery overnight and run exclusively on battery power for the first 30 miles or so before you have to start using gas. Since most people commute less than 30 miles to work, that sort of thing could have a huge impact on oil consumption if it can be done inexpensively enough. The economics of hybrids will undoubtedly look a whole lot better if gas is sitting at five dollars per gallon, and seem likely to stay there.

Of course the electricity for plug-in hybrids has to be generated somewhere, but our sources can be much more flexible there. Interest in nuclear power seems to be increasing again. Solar cells are finally starting to get close to competitive economically. What's the bottom line on all of this? Oil prices will probably trend up in spite of decreased demand from the US and Western Europe, but in the long run that will help us make a much needed transition, first to more efficient use of energy, and then to newer, more sustainable sources of energy.