In a recent editorial the WSJ argued (sub req; excerpt via Lynne Kiesling):

There's been unconscionable behavior all right, most of it on Capitol Hill. A decent portion of the latest run-up in gas prices -- and the entire cause of recent spot shortages -- is the direct result of the energy bill Congress passed last summer. That self-serving legislation handed Congress's friends in the ethanol lobby a mandate that forces drivers to use 7.5 billion gallons annually of that oxygenate by 2012.

At the same time, Congress refused to provide liability protection to the makers of MTBE, a rival oxygenate getting hit with lawsuits. So MTBE makers are leaving the market in a rush, while overstretched ethanol producers (despite their promises) are in no way equipped to compensate for the loss of MTBE in the fuel supply. Ethanol is also difficult to ship and store outside of the Midwest, which is causing supply headaches and spot gas shortages along the East Coast and Texas.

Sounds like a testable hypothesis so yours truly tracked down gas prices and states' distances from Iowa (measured state capital to state capital). Consistent with WSJ's argument about supply disruptions in Texas and the East, I included the 37 states lying east of an imaginary line extending from Texas to North Dakota.

It seems the WSJ may be onto something*--the correlation between the change in gas prices over the March 25-April 25 period and the distance from Iowa is 0.65. States more distant from Iowa have experienced larger gas price increases over the past month.

I then turned to regression analysis to estimate the magnitude of the effect--roughly 0.72 cents per additional 100 miles of distance from Iowa. For example, Atlanta is 971 miles from Des Moines so the estimate predicts GA gas should have increased about 7 cents per gallon more than IA gas. Turns out it did--GA gas prices went up 45 cents over the month while IA gas prices increased 38 cents.

Bottom line: The grandstanding clowns in DC do indeed deserve some of the blame for the current gas price spike. But look again at the GA-IA comparison in the previous paragraph. Iowa's price hike is far larger than the extra bump that GA has gotten courtesy of the ethanol rent seekers. I, therefore, conclude that $70+ oil is more responsible for current gas prices than ethanol. Of course, our ethanol policy is inexcusable even if it isn't the primary culprit in the gas price increase.

*See also this post in which Lynne Kiesling explains the economics of the ethanol/MTBE mess.