Popping Corn

Every time we post something on this blog about biofuels (such as ethanol and biodiesel), we get a bunch of, shall we say, spirited comments.  Passions run hot on both sides, with opinions split between those who think that biofuels are one of the most promising solutions to America's petroleum dependence and a great way of reducing climate-warming emissions, and those who think that that biofuels are mostly a costly and wasteful distraction.

I'm not quite sure what to think.  Clearly, this is an area where people's opinions, hopes, and even values matter much, much less than the actual facts.  But when it comes to biofuels, there seem to be a lot of basic disagreements about core facts--a problem that is compounded by the fact that many parties to the debate have strong opinions about which facts ought to be true.

Still, amid the swirl of competing factual claims and counterclaims, two things are abundantly clear. First, making transportation fuel from living matter involves complicated tradeoffs; the answer to whether it's a good thing isn't always apparent, and certainly isn't always what you think at first glance.  And second, when you look at the specifics, some biofuels lose their luster.

And nowhere are these points more evident than in the hoopla over producing ethanol--a gasoline substitute--from corn.

The corn ethanol debate tends to get detailed and technical very quickly.  But the details really matter, so I'll try to summarize them as best I understand them.

The most-cited critic of corn-based ethanol, Dr. David Pimentel from Cornell, claims that ethanol has no fossil fuel benefits whatsoever.  That is, it takes more fossil fuel energy--in the form of tractor fuel, nitrogen fertilizers, electricity to run ethanol plants, and so on--to make a gallon of ethanol than is contained in the gasoline that the ethanol displaces. (See, e.g., this pdf.) If this claim is true, then converting corn to ethanol is really a waste of fossil fuels, rather than a way to reduce fossil fuel use.

But Pimentel's critics (and there are plenty) say the figures he uses are simply outdated. Corn yields per acre are rising, the energy cost of making fertilizers is falling, and ethanol plants have gotten more efficient.  By using outdated figures, the critics say, Pimentel overestimates how much fossil energy it takes to make a gallon of ethanol. 

This criticism of Pimentel makes sense to me, and given the unanimity of the critics I have little reason to doubt that it's correct. 

On the other side, the most frequently cited pro-ethanol researcher is USDA's Hosein Shapouri, who argues that corn ethanol's "net energy balance"--the energy contained in ethanol, minus the amount of fossil energy required to make it--is positive and growing. That is, you get a lot more energy out than you put in. (See, e.g., this pdf).  But reading over Shapouri's analysis, it seems just as flawed as Pimentel's. 

First, Shapouri ignores some of the energy costs of ethanol.  His analysis doesn't account for the energy required to manufacture farm machinery, to build ethanol plants, or to irrigate corn. These are real energy costs, but energy analysts typically ignore them.  (Pimentel, to his credit, does include estimates for all of these steps -- though once again his critics say that his numbers are out of date.)

Second, Shapouri says that a lot of the energy used to produce ethanol should really be attributed to the byproducts of ethanol production.  Shapouri estimates that corn starch -- which is used to produce ethanol -- is only about two thirds of the weight of a kernel of corn.  The other third of the kernel goes into the byproducts, such as corn gluten and corn oil. When you apportion one-third of the energy cost of ethanol distillation to the byproducts, rather than ethanol itself, the net energy balance of ethanol starts to look a lot better.

But the problem here is that bulk of the byproducts is simply animal feed, just as corn is animal feed.  It seems odd to attribute so much energy--a third of the total used to convert corn to ethanol--to converting one form of animal feed into another. So Shapouri's treatment of the byproducts may give an unduly rosy view of ethanol's energy balance.

And finally, Shapouri (like Pimentel) assumes industry-wide averages for most of his numbers -- that is, average US corn yields, average fertilizer inputs, average energy inputs to ethanol plants.  But as Shapouri's own paper shows, there's a huge difference in crop yields and fertilizer usage just among the midwestern states; Iowa's yields are about 50% higher than South Dakota's.  Just so, there are huge differences in cropland yields within each state; some cropland is especially prone to drought or flooding, and is simply less productive than the best farmland. 

This makes me think that the right way to do the analysis is to look at the land that would go out of production if it weren't for ethanol subsidies.  That's the land that's likely to have the lowest yields and the highest input costs--which, on average, will be the land with the worst net energy balance.

So the bottom line, as best I can tell, is this.  Pimentel uses outdated numbers.  Shapouri ignores some of the energy costs of producing corn and ethanol, and may give too much credit to byproducts.  And both of them use national averages for yields and inputs, rather than the rates for marginal cropland.

All of which makes me think that reality is probably somewhere between the two studies:  namely, that -- given today's yields and inputs -- corn-ethanol may produce either a small net energy gain, or a small loss, compared with the amount of fossil fuels used in production.  But mostly it's a way of converting natural gas and coal into liquid transportation fuel. 

Of course, if you're trying to reduce petroleum imports, that may not be such a bad thing.  North America's natural gas supplies are dwindling, but at least it still has a lot of coal.  So corn ethanol may reduce American energy imports.

Except it's not clear that using ethanol actually decreases petroleum consumption all that much.  Increasing the supply of transportation fuels tends to decrease the price; decreasing the price tends to increase demand.  Estimates of the size of this effect vary (see here for a review), but a rule of thumb seems to be that when gas prices fall by 10 percent over the long term, gas consumption rises by about 6 percent.  Which means that adding more ethanol to the fuel supply may do less to reduce petroleum consumption than one might hope.

Two other points worth mentioning.  First, producing corn has substantial environmental costs, ranging from water quality degradation (e.g., the dead zone in the Gulf of Mexico) to habitat loss.  Those ought to be considered in any environmental accounting of corn ethanol, just as the many externalities of gasoline--spills, drilling for oil in the last great places, etc.--should be considered.

Second, let's assume that one of the chief reasons to promote ethanol is to reduce global warming.  Growing corn using no-till farming techniques does sequester carbon, about .1 to .2 tons per acre per year by some estimates.  But restoring cropland to native prairie sequesters even more -- by some estimates, over a ton per acre per year (see here for example).  Plus, the use of nitrogen fertilizers to grow corn can release significant amounts of nitrous oxides, a potent global warming gas.  (See p. 47 of this pdf, which also has a great overview of the various corn ethanol studies.)  So even if ethanol has a slight net positive energy balance, it may still not be a particularly good choice from a greenhouse gas perspective.

I'll leave aside, for the moment, the argument that corn ethanol is really the first step in a transition to more benign forms of biofuels.  But suffice it to say, a close look at these issues has convinced me that getting excited, one way or another, about corn ethanol is mostly a distraction from far more important tasks -- namely, encouraging vast improvements in the efficiency of the vehicle fleet, along with changes to urban design that make driving less necessary.

Those have potential to substantially cut back petroleum imports, CO2 emissions, and gasoline consumption.  Corn ethanol, at least for the near future, is simply beside the point.