If This Is Peak Oil, Then I’m Not Sure What The Problem Is

A guest poster over at Econbrowser claims that the new EIA oil supply projections are a “Hard Core Peak Oil Forecast”. His argument is that the EIA has been steadily reducing its oil supply projections for 2020 over its last four annual projections, and that this is a sign that EIA has “placed its fortunes firmly with the peak oil crowd”. I think he’s wrong. In any case, if this is what counts as peak oil, then I’m not sure why anyone is so excited about it.

Here’s the graph that attempts to make the case:

Sure enough, EIA has been steadily reducing its projections for 2020, from 104 million barrels per day (mb/d) in its 2007 projection to 92 mb/d in its most recent projections (released last month).

But why have the projections gone down? It’s possible to more than explain the whole difference through revised expectations for economic growth. The decreased projections for oil supply have almost nothing to do with changed beliefs about the prospects for oil supply development. They are pretty much entirely explained by external factors.

The 2007 projections pegged total world GDP in 2020 at 107.1 trillion (2000) dollars, which is equivalent (XLS file) to about 118 trillion (2005) dollars. (I’ve used a “world” GDP deflator, rather than a U.S. one, to convert from 2000 to 2005; switching measures doesn’t make a big difference.) The 2010 EIA projections expect GDP in 2020 to be 97.5 trillion (2005) dollars, for a 17% reduction in expected GDP. Contrast that with a 13% reduction in expected oil production. If the GDP-elasticity of oil demand were 0.64, the reduced GDP expectations would fully explain the lower oil production estimates. As is stands, long-run income-elasticity of oil demand is almost certainly higher than that, so revised GDP estimates more than explain the lower supply projections. Indeed the interesting puzzle may be why oil supply is so high in the new estimate, rather than why it is so low.

The only way out for someone who wants to blame the lower supply projections on peak oil is to argue that the lower GDP projections are themselves a consequence of peak oil. The argument would say that the lower GDP projections are the result of the current recession (true), that the current recession was the result of high oil prices (possible [PDF] but debatable [PDF]), and that the high oil prices of 2002-2008 were the result of peak oil (who the heck knows). But that is not the case that the EIA is making: It projects oil prices to rise to about $110/bbl in 2020, and to over $180/bbl [PDF] in its “high oil price case”, but in neither case does it project a recession. Indeed the impact of the high oil price case on economic growth is tiny.

Moreover, if $110/bbl oil counts as “peak oil” (since, after all, we’ve agreed to use the EIA forecasts), then I’m not sure what anyone’s so worried about. Are people really claiming that $110/bbl oil will be catastrophic?

The EIA may be wrong about how well economies can withstand sustained high oil prices and about how much oil the world can produce at a given price if called on. But it is clearly not in the peak oil camp.