Media & News

Making Mountains Out of Molehills

November 8, 2010


Last week, NRDC's Nathanael Greene noted my frustration with his organization's hasty dismissal of new Department of Energy research that shows "...minimal to zero indirect land use change was induced by use of corn for ethanol over the last decade." NRDC immediately attempted to trash the DOE work, which was conducted by seasoned scientists at the agency's Oak Ridge National Laboratory. Nathanael went so far as to suggest it "tells us nothing about ILUC." It's interesting that NRDC could reach such a conclusion without actually ever having seen the study (it hasn't been published yet), without ever having laid eyes on the raw data underlying the analysis, or without ever having experimented with the methodology that ORNL used. Rather, they summarily dismissed the ORNL research as "one-sided" and "flawed" based on a brief PowerPoint presentation that describes the findings and the basic analytical approach. I could be wrong, but I doubt NRDC took the time to reach out to ORNL to seek further explanation or ask any questions. And I doubt NRDC took the time to learn more about the index decomposition analysis methodology used by ORNL. NRDC's beef with the ORNL work is that, in Nathanael's words, "It doesn't even attempt to compare land-use with the policies to what would have happened without the policies." This is a familiar argument from Nathanael, and it has been the default response from NRDC every time someone has challenged ILUC with empirical data. (Note: After learning a little more about the ORNL analysis, I'm not convinced that NRDC's "lack-of-a-baseline" argument is entirely valid—the systematic decomposition analytical method used by the researchers allows them to isolate the contributions to LUC of individual demand drivers, like ethanol demand, while holding other factors constant). So, yes, I was a little frustrated. OK, I was a lot frustrated. I was (and am) frustrated not only by NRDC's half-baked and knee-jerk rejection of the ORNL findings, but also by the organization's recent failure to contribute anything constructive to the debate over ILUC.  I'm frustrated that there's no way to validate the actual significance or magnitude of ILUC. And I'm frustrated that every time someone tries to "ground truth" the ILUC theory using real data, they get trashed by NRDC. Most of all, I'm frustrated that NRDC is stuck in first gear on ILUC, while the rest of us have at least shifted into second. While academia, government and the biofuels industry are continuing their efforts to improve the understanding of ILUC, collect more and better data, and refine analytical methods, NRDC and other ENGOs are standing idly by waiting to blast any new information or research that disagrees with their pre-conceived notions. They're still clinging to the discredited findings of Timothy Searchinger, while the rest of the world has moved on to better methodologies and better data. In responding to comments by RFA's Matt Hartwig about NRDC's abrupt reversal of positions on ethanol, Nathanael stated that "following the science" on ILUC caused him and others to change their minds about ethanol. Well, it turns out they only followed the science so far—they followed it as far as February 2008, to be exact, which is when Searchinger's paper was published. They've been following something else since then (namely the anti-biofuels rhetoric of the Grocery Manufacturer's Association and oil refiners). If NRDC had continued to "follow the science" on ILUC with the rest of us, they'd recognize that the latest research on ILUC shows the early estimates of Searchinger and others were likely overblown by a factor of at least seven (Tyner et al.) or more (ORNL). But I digress. When I set out to write this, I was intending to address some inferences about ILUC that Nathanael made in his last post based on some USDA corn export data. Nathanael suggested that ILUC is occurring globally because U.S. corn exports constitute a smaller percentage of world corn exports today than in 1990. And though he doesn't come right out and say it, his insinuation is that U.S. corn exports constitute a smaller share of the global trade today because we're using more for ethanol. Ergo, according to his logic, using more corn for ethanol in the U.S. has reduced the U.S. share of global corn exports, which is leading to LUC in foreign countries. This assertion has some obvious holes. First off, the chart tells us nothing about cause and effect. Nathanael is implying that using more corn for ethanol in the U.S. is the cause and the effect is that the U.S. has a smaller percent share of the global corn trade than it did in 1990. But how does he know it isn't the other way around? That is, it could very well be that the cause is increased foreign corn production and exports and the effect is a smaller percent share of the global corn trade for the U.S. Second, the chart blatantly omits exports of feed co-products from ethanol production, such as distillers grains (DG) and corn gluten feed (CGF). This is a glaring omission because these co-products replace corn (and some soybean meal) in the international feed market and effectively reduce the need for some amount of U.S. corn and soymeal exports. The U.S. will export approximately 10 million metric tons of DG and CGF this year, meaning total corn and corn co-product exports will likely top 60 million metric tons. When co-products are added to Nathanael's chart, the current U.S. percent share of world corn and corn co-product exports rises from 54% to about 60%. And, more important, when co-products are added, the U.S. percent share of the global trade is generally steady from 1999 to 2010, i.e. the "biofuels era." Would U.S. corn exports have constituted a larger percent share of the global trade if ethanol didn't exist? Well, no one knows for sure. But based on the "pre-biofuels era" trend...the answer would be "No." As Nathanael's chart clearly shows, the U.S. share of world corn exports was trending downward long before ethanol came into play. Why should we believe the trend would have reversed if we didn't have ethanol (more on that in a moment)? Third, why would it be a bad thing for foreign countries to become more adept at feeding themselves, and why is it bad that other nations are producing enough surplus to actually have some to export? Remember the old Chinese proverb? "Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime." Well, that's what's happening in the global grain market. Advances in technology and positive price signals are allowing farmers around the globe to profitably produce more grain on virtually the same acreage. It wasn't so long ago that the U.S. corn industry was being accused of "flooding the world market" with cheap corn and putting economically disadvantaged farmers in other nations out of business. Now that those farmers in other nations are earning a decent return on their crops, U.S. farmers are being criticized for providing a smaller percentage of global corn exports than they did in the 1990s era of $2 corn. You simply can't have it both ways! But the most important question about Nathanael's corn export chart is: what does it really tell us about the impact of U.S. ethanol growth on U.S. exports of corn? After all, that's what we're trying to figure out, right? And after all, that's what drives the entire ILUC calculus in the economic models being used by CARB and EPA. Sadly, the chart does nothing to get us any closer to answering that question. Sure, it shows that the U.S. share of the world corn trade has been essentially flat since 2000 (when co-product exports are considered), but that sure as heck doesn't mean it would have been any different without ethanol. Let me illustrate. The chart below shows corn & corn co-product exports and soybean & soybean meal exports in the "pre-biofuels era" compared to the "biofuels era." The chart also shows the "pre-biofuels" trend, which could be thought of as a baseline of sorts to indicate a world without biofuels. One can plainly see that in the "pre-biofuels era" (1975-1998), U.S. corn exports were actually trending downward slightly. U.S. soybean & soybean meal exports were trending upward slightly. So, one could argue that in the absence of biofuels, exports of corn and exports of soybeans & soybean meal might very well have continued along their historical trend lines. But in the biofuels era (1999-2010), exports of both corn & corn co-products and soybeans & soybean meal have deviated significantly from the historical "pre-biofuels" trend. In both cases, exports have been higher than the historical "pre-biofuels" trend line suggests might have otherwise been the case. So, how can someone argue with a straight face that biofuels growth in the U.S. has reduced exports of corn & co-products and soybeans & soybean meal? Few advocates of the ILUC theory understand that U.S. grain exports aren't well correlated to global population growth (in fact, there is very little correlation to GDP growth either). You can't simply assume that just because the world population is growing, that U.S. grain exports will grow along at a tandem rate. It hasn't ever worked that way historically and there's no reason to believe it will change, whether we're producing biofuels or not. NRDC has criticized the ORNL work and the other empirically-focused analyses for purportedly not referencing a baseline scenario or not examining what would the world might have looked like without biofuels. As I expressed in my response to Nathanael's post, we don't have the luxury of knowing with certainty or precision what the world would look like without biofuels. But we can learn from the past, and there is a strong argument to be made that trending forward the events of the "pre-biofuels era" could serve as a decent proxy for a "without biofuels" baseline scenario. That is, in many ways, what the ORNL analysis does. In attempting to justify its reflexive repudiation of the ORNL work, NRDC has been using a strange little analogy about an airplane preparing to fly over a mountain. They say simply pointing to rising U.S. corn and soy exports as evidence that ILUC is minimal is akin to the pilot assuming his plane will get over the mountaintop just because the nose is pointed upward. But that grossly oversimplifies what ORNL is really saying and misses the point on many levels. In essence, ORNL is saying that "based on where we've been, and based on our trajectory, we're going to clear the mountain, which, as it turns out, may actually be more of a molehill anyway." In any case, who would you believe? The seasoned pilot with years of experience and the best available instrumentation, or the hysterical aviophobic passenger sitting next to you who insists that the plane is absolutely going to crash into the mountain? (Indeed, the current regulatory practice of assigning an ILUC adder to the lifecycle carbon intensity score of biofuels implicitly assumes that the plane will absolutely crash, when history shows us the chances are far better that it clear will the mountain). I'll trust the pilot every time.