I was thinking about doing an economic analysis of reductions in oil consumption today and happened to receive the latest issue of Resources, a magazine put out by the US-based Resources For the Future institute, probably the leading advocate of market-based solutions to environmental problems. It included an article on the topic, Is Gasoline Undertaxed in the United States (Note: PDF file). The conclusion, not surprisingly is “Yes”, basically because of congestion, pollution and accident “externalities”. The author, Ian Parry, suggests an optimal tax of $1/gallon compared to the current tax of about 40 cents/gallon.
Assuming current demand is based on a ‘normal’ price of around $1.20/gallon, the implied price increase is 50 per cent. With a long-run elasticity of demand of 0.7, this implies that the long-run reduction in demand would be 35 per cent. The implication of Parry’s analysis is that the US, acting in its own self-interest, and regardless of foreign policy concerns, would be better off reducing gasoline consumption by 35 per cent. Broadly similar arguments apply to other transportation uses. Since transport accounts for about 60 per cent of total petroleum consumption, this implies a reduction in total oil consumption of 20 per cent, and a reduction in net imports of around 40 per cent or about 4 million barrels per day. This is more than the likely amount of additional production that would arise from overthrowing Saddam, even assuming an American occupation or an Iraqi puppet government acting in the interests of the US, as is explicit in the arguments of Steven den Beste and implicit in any oil-based scenario.
Rationing would achieve the same reduction in demand more rapidly, although less efficiently in an economic sense. If reducing reliance on Saudi oil is a strategic objective of the US of sufficient importance to justify war, then the US government is morally obliged to take whatever peaceful and unilateral action it can towards this goal before resorting to military means.

In the comments thread, Steven den Beste asserts that “Your entire idea is based on the fallacious assumption that the refining process can be adjusted to produce all outputs in whatever proportions are desired. That isn’t how it works. ”

Actually, the refining process can be adjusted, though not without limit, and, more importantly, there is trade in petroleum products as well as crude. The US uses more gasoline relative to other products than the fractionation process produces, and therefore trades other refined products for gasoline as well as importing crude. As this graph shows, transportation (mainly gasoline) is indeed the dominant use of petroleum in the US. In any case, I picked gasoline precisely because this is the petroleum product for which attempts to reduce consumption would be most politically unthinkable. If the US government could grasp this nettle, it would have no trouble in cutting back on other uses for oil.

The idea of a war based even partly on promoting a flow of Iraqi oil doesn’t stand up either economically or morally. A war with Iraq may be justified for a number of reasons, but oil isn’t one of them.