I admit to a certain fascination with the informal “law of unintended consequences.” I think this is a very useful concept and should always be kept in mind. Essentially this axiom holds that the world is a complex place with far more variables in play than we could ever anticipate. Therefore our actions are likely to have consequences that we did not intend.

It therefore makes sense to be cautious and humble when taking any big action (like passing a law, establishing a regulation, promoting an industry, etc.), and to build into the process monitoring the effects of the intervention and making course corrections as necessary.

A new study suggests that biofuels may be the latest entry in the book of unintended consequences. The study tries to address a very narrow question – how much of the carbon that is released into the atmosphere when biofuels are burned was offset by the taking up of carbon from the atmosphere when the plants used to create the biofuels (feedstock) were grown?

Prior analysis of the utility of biofuels assumes that the offset is 100%, that all of the carbon released when biofuels are burned came from the atmosphere when the plants were grown. One type of analysis, called a lifecycle analysis (LCA), makes this assumption, and therefore focuses on the energy used in the entire lifecyle of creating and transporting biofuels. In an LCA the question is, how much energy is put into the biofuels vs how much energy do we get out, assuming the biofuel itself is carbon neutral.

There is an ongoing debate about how useful LCAs are, and if they are the best analysis or if they can be misleading. In any case, if you asked the question about the net energy balance of biofuels 20 years ago the answer was probably that biofuels cost more energy to create than they provided. This, of course, depends on the feedstock. That was the answer for corn ethanol, but sugarcane biofuels were likely efficient due to the higher energy density of the plants.

Improvements in efficiency over the years have slowly shifted the results of LCAs and recently it seems that biofuels are reasonably energy efficient – they produce more energy than they consume. There is still debate about this, but that is not the focus of the current study.

This new study asks a different question: is the assumption of carbon neutrality in the growing and burning of biofuels justified? You could also frame this in terms of carbon offset. Current LCAs assume that the carbon offset of growing the feedstock is 100%, that the crops pull enough carbon out of the air to 100% offset the carbon released during burning of the biofuel. But what if this assumption is incorrect?

The authors essentially analysed the increase in carbon capture by farms growing feedstock for biofuels from 2005-2013 and compared that to the amount of biofuel created and the amount of CO2 that would release. (You can read the paper for all the technical details, I don’t think I can add anything by trying to summarize it here.)

What they found is that, rather than being 100%, the offset of growing feedstock for biofuels is only 37%. They modeled the entire carbon flow through the system, including material carbon and carbon from fossil fuel-based fertilizers. They did not account for the effects of increased land use for growing feedstock for fossil fuel.

When they plug their 37% figure into current LCAs for biofuel efficiency they find that the use of biofuels has resulted in a net increase in carbon release, and in fact is worse than just burning fossil fuels. Ouch!

They conclude:

The assumption that biofuels are inherently carbon neutral is a premise of most climate-related fuel policies promulgated to date, including measures such as the LCFS and RFS that evaluate GHG impacts using lifecycle modeling. However, this analysis found that the gains in CO₂ uptake by feedstock were enough to offset biofuel-related biogenic CO₂ emissions by only 37 % over 2005–2013, showing that biofuel use fell well short of being carbon neutral even before considering process emissions.

No one study is ever going to be the final word on a complex question like the net effects of using biofuels. This is also a moving target as production methods are slowly improving over time. Further, researchers are looking into using different feedstocks and different methods of production. This analysis, for example, would not apply if we figure out a way of using algae for feedstock and bacteria to convert it to biofuels.

The study, however, raises some very serious questions about dedicating farmland to growing corn for ethanol for biofuels. It seems that this may be a net negative in terms of carbon release, even if it has tipped over into being a net positive for energy itself.