For the past few years, the Renewable Fuels Association (RFA) has been claiming that U.S. production of corn ethanol has had massive benefits for consumers by lowering the cost of gasoline at the pump by around $1 per gallon. If you live outside of D.C. you may not have heard about this, but near the Capitol you can’t turn around without seeing one of these.
The RFA isn’t arguing that U.S. ethanol is lowering the world price of crude oil. Rather, they claim that without corn ethanol, refinery margins – the difference between the crude price and the wholesale gasoline price — would be higher than they have ever been in any of the last 30 years. Margins would be at levels that would yield gigantic profits for any refiner in operation. And yet, the RFA is implicitly arguing, there would be no new entry into the refining market, no expansion of existing refining facilities and no increased imports of gasoline (even though gasoline would be much more valuable in the U.S. than in the rest of the world). And the closures of older refineries would have occurred at the same pace as they have, even though those refineries would have been earning operating margins more than three times higher than their average in the last decade. The economics of the RFA claim has never made sense.
Now, Professors Christopher Knittel (MIT) and Aaron Smith (UC Davis) have shown that the reason the economic claims don’t make sense is that the claims are based on a research paper in which the statistics don’t make sense. In a new working paper of the UC Center for Energy and Environmental Economics, Knittel and Smith point out a number of flaws with the original 2009 study and the numerous follow-ups that the RFA has funded by two professors, at Iowa State and Wisconsin. [Full Disclosure: I was Knittel’s dissertation advisor in the 1990s and have co-authored a couple papers with him on electricity markets.]
Basically, the original study found that ethanol production has gone up since 2000 at a time when oil prices skyrocketed so the refinery margin share of the gasoline price has declined. From this, the authors jumped to the conclusion that increased ethanol output has caused refinery margins to fall. Knittel and Smith show that reasonable changes in the statistical analysis – such as recognizing that refinery margins generally move inversely with the price of oil — causes the results to disappear or drastically shrink. They also show that the “impact’’ the original paper claims to show is simply the negative correlation between ethanol output over the last decade and the particular way the original paper constructed refinery markups. Knittel and Smith show that the same flawed logic could show significant “impact’’ of ethanol production on the price of natural gas (negative), the unemployment rate (positive), and even the ages of the Knittel and Smith’s children (positive). In other words, ethanol production has increased in the last decade; anything that has systematically decreased or increased over this time will lead to the false conclusion of causality by this logic.
The RFA claim has not been taken seriously among energy economists, but it has in D.C. Secretary of Agriculture Vilsack has quoted it in his speeches. Knittel and Smith are not going to make their careers with this paper, but it is a huge contribution to the public policy debate. They deserve recognition for doing the detailed analysis to fact check the RFA advocacy and the research on which it is based. I hope it will drive a stake through this one nonsense claim in the energy debate. Unfortunately, there are plenty of others – coming from both ends of the political spectrum – that continue to undermine attempts at reasoned debate.