Institute of Transportation Studies
UC Berkeley Transportation
Sustainability Research Center
( University of California, Berkeley)
Year 2007 Paper UCB - ITS - TSRC - RR - 2007 - 1
Creating Markets for Green Biofuels:
Measuring and improving environmental
performance
Brian T. Turner  Richard J. Plevin†
Michael O’Hare‡ Alexander E. Farrell  
 Energy and Resources Group, Richard and Rhoda Goldman School of Public Policy, UC
Berkeley
† Energy and Resources Group, UC Berkeley
‡ Richard and Rhoda Goldman School of Public Policy, UC Berkeley
  Energy and Resources Group, UC Berkeley Transportation Sustainability Research Center
This paper is posted at the eScholarship Repository, University of California.
http:// repositories. cdlib. org/ its/ tsrc/ UCB- ITS- TSRC- RR- 2007- 1
Copyright  c 2007 by the authors.
Creating Markets for Green Biofuels:
Measuring and improving environmental
performance
Abstract
This study describes how some biofuels are produced, emphasizing agricul-tural
production systems, and considers what is needed in order to measure
and communicate environmental performance, and gives examples of how this
might be done. We describe a set of seven uses of a Green Biofuels Index, from
a wholly market- driven implementation through a set of increasingly intrusive
regulatory approaches. We then present several case studies of specific biofuel
production pathways using a lifecycle analysis of the inputs to feedstock pro-duction
and processing, but excluding market- mediated effects.
We recommend four steps to create markets for green biofuels: 1. Measure
the global warming intensity of biofuels. 2. Measure the overall environmental
performance of biomass feedstock production. 3. Develop and implement a
combined Green Biofuels Index. 4. Research better practices, assessment tools,
and assurance methods.
Creating Markets for Green Biofuels:
Measuring and improving environmental performance
Brian T. Turner1,2, Richard J. Plevin1, Michael O’Hare2,
Alexander E. Farrell1,3
RESEARCH REPORT
UCB- ITS- TSRC- RR- 2007- 1
April 2007
1Energy and Resources Group, UC Berkeley
2Richard and Rhoda Goldman School of Public Policy
3UC Berkeley Transportation Sustainability Research Center
The Transportation Sustainability Research Center fosters research,
education, and outreach so that transportation can serve to improve
economic growth, environmental quality and equity. It is housed at
the UC Berkeley Institute of Transportation Studies.
http:// www. its. berkeley. edu/ sustainabilitycenter/
Creating Markets for Green Biofuels
Brian T. Turner1,2, Richard J. Plevin1, Michael O’Hare2, Alexander E. Farrell1
1Energy and Resources Group, UC Berkeley; 2Richard & Rhoda Goldman School of Public Policy, UC Berkeley
This research was partially supported by the Natural Resources Defense Council, National Science
Foundation's Climate Decision Making Center at Carnegie Mellon University ( SES- 034578) ( to Farrell) and
Graduate Research Fellowship program ( to Jones and Plevin) and the Goldman School of Public Policy ( to
O’Hare). However the view and opinions herein, as well as any remaining errors, are those of the authors
alone and do not necessarily represent the views of the sponsors.
Creating Markets For Green Biofuels
ii
Table of contents
Executive Summary............................................................................................................ v
1 Introduction .................................................................................................................. 1
2 How Biofuels Are Produced......................................................................................... 4
2.1 Feedstock production.......................................................................................... 5
2.2 Fuel processing................................................................................................... 6
2.3 Environmental consequences ............................................................................. 8
3 Measuring and Communicating Environmental Performance ................................... 11
3.1 Feedstock.......................................................................................................... 11
3.2 Processing......................................................................................................... 14
3.3 Performance- vs. practice- based standards....................................................... 15
3.4 Quantitative GHG measurement....................................................................... 16
3.5 Leakage............................................................................................................. 16
3.6 Practical implementation .................................................................................. 17
3.7 Dimensionality ................................................................................................. 19
3.8 Compensatory vs. mandatory minimums ......................................................... 20
3.9 Tracking, trading, and banking......................................................................... 20
3.10 Compatibility with other regulatory structures................................................. 22
3.11 International trade............................................................................................. 22
3.12 Legibility, convention, and implicit ceilings .................................................... 23
4 Examples of green biofuels indices............................................................................ 24
4.1 A quantitative compensatory index .................................................................. 24
4.2 A qualitative compensatory index .................................................................... 27
4.3 A lexicographic non- compensatory index........................................................ 28
4.4 Blending fuels and feedstocks .......................................................................... 29
5 Implementations of a Green Biofuels Index............................................................... 30
5.1 Allow the market to find its way to efficient labeling and claims.................... 30
5.2 Define allowable claims and protocols to support them................................... 30
5.3 Require environmental labeling........................................................................ 31
5.4 Require government ( and contractors) to purchase green biofuels .................. 32
5.5 Subsidize or tax based on environmental performance .................................... 32
5.6 Require an aggregate green biofuels performance ........................................... 33
5.7 Forbid sale of fuel below some level................................................................ 34
6 Case Studies ............................................................................................................... 35
6.1 Feedstock production........................................................................................ 35
6.2 Biorefining........................................................................................................ 36
6.3 Corn ethanol 1: Coal- fired ethanol production with cogenerated
electricity 37
6.4 Corn ethanol 2: Natural gas– fired ethanol production...................................... 38
6.5 Corn ethanol 3: Integrated ethanol production / animal feedlot ....................... 38
6.6 Corn ethanol 4: Biomass- powered ethanol production..................................... 39
6.7 Cellulosic ethanol production........................................................................... 40
6.8 Case studies summary ...................................................................................... 40
7 Recommendations ...................................................................................................... 42
8 References .................................................................................................................. 44
Appendix A: Measuring Multiple Dimensions of Environmental Performance………… 47
Appendix B: Other Certification Systems .…………….……….……….……….……... 53
Appendix C: Lifecycle Assessment of Biofuels……….……….……….……….……… 57
Creating Markets For Green Biofuels
iii
Figures
Figure 1: General Biofuel Pathway With Inputs and Environmental Impacts……………. 11
Figure 2: Key U. S. Energy Crop Production Pathways…………………………………… 15
Figure 3: A Qualitative Compensatory Green Biofuels Index ( Illustrative) ...……………. 35
Figure 4. Well- to- Tank GHG Emissions From Various Fuel Pathways ( Illustrative)…….. 48
Figure C- 1. Corn Ethanol Production Lifecycle…………………………………………… 58
Tables
Table 1: A Quantitative Compensatory Green Biofuels Index …….…………………..… 33
Table 2: A Lexicographic ( Non- Compensatory) Green Biofuels Index…………….……. 36
Creating Markets For Green Biofuels
iv
Abbreviations Used
CHP Combined heat and power
CRP Conservation Reserve Program
CSP Conservation Security Program
DG Distillers grains
DDGS Dried distillers grains with solubles
EBAMM ERG Biomass Analysis Meta Model
FSC Forest Stewardship Council
GGE Gasoline gallon equivalent
GHG Greenhouse gas
GMO Genetically modified organism
GREET Greenhouse Gas, Regulated Emissions and Energy Use in
Transportation ( well- to- wheels LCA model)
GWI Global warming intensity
GWP Global warming potential
IATP Institute of Agriculture and Trade Policy
LCA Lifecycle assessment
LEED Leadership in Energy and Environmental Design
MJ Megajoule
MSW Municipal solid waste
NIST National Institute of Standards and Technology
NOP National Organic Program
RIN Renewable Identification Number
RFS Renewable Fuel Standard
SRWC Short- rotation woody crops
USDA U. S. Department of Agriculture
VOC Volatile organic compound
WDG Wet distillers grains
Creating Markets For Green Biofuels
v
Executive Summary
While “ green” and “ environmentally friendly” may seem synonymous with “ biofuels,” this is not
necessarily true in practice; all biofuels entail tradeoffs among positive and negative
environmental effects. Because the environmental performance of biofuels is not measured
today, consumers have no information about how to buy greener biofuels and producers have no
incentive to manufacture and market them. The right set of market signals and regulatory
requirements can change this situation, so that American consumers could buy biofuels certified
as environmentally friendly, and so that the American agriculture and energy industries would
have incentives to improve. Markets for green biofuels would stimulate a new wave of
innovation, creating high- value and truly green biofuels, and enhancing energy security by
diversifying our energy sources. However, without appropriate information, incentives, and
rules, the biofuels industry is likely to expand production in environmentally harmful ways.
This study describes how some biofuels are produced, emphasizing agricultural production
systems, and considers what is needed in order to measure and communicate environmental
performance, and gives examples of how this might be done. We describe a set of seven uses of a
Green Biofuels Index, from a wholly market- driven implementation through a set of increasingly
intrusive regulatory approaches. We then present several case studies of specific biofuel
production pathways using a lifecycle analysis of the inputs to feedstock production and
processing, but excluding market- mediated effects.
We recommend four steps to create markets for green biofuels:
1. Measure the global warming intensity of biofuels.
2. Measure the overall environmental performance of biomass feedstock production.
3. Develop and implement a combined Green Biofuels Index.
4. Research better practices, assessment tools, and assurance methods.
A Green Biofuels Index should be developed through a cooperative effort by environmental and
energy regulators, agricultural agencies, and stakeholders from these communities, at either the
state or national levels. Such an approach could be extended to other fuels as well.
Research is needed to develop better methods for producing biofuels as well as better ways of
assessing and verifying the environmental performance of biofuels. Further work is crucially
needed to address uncertainties and missing elements in current approaches, especially in
agricultural greenhouse gas emissions, the effects of land- use change, greenhouse gas emission
and ecosystems impacts associated with biomass thinning in forests, and indirect effects created
by changes in markets for biomass feedstocks or food. The National Academies could, along
with appropriate scientific bodies and stakeholders, help identify a research agenda to enable and
expand markets for green biofuels. Regulators, the National Science Foundation, and other
appropriate agencies ( federal, state, and private) should support such a research agenda.
Creating Markets For Green Biofuels
vi
“ Mom, why are we getting gas from this green pump?”
“ Because this is where we get ethanol instead of gasoline.”
“ Why don’t we use gasoline?”
“ Gasoline comes from oil in the ground and puts carbon dioxide in the air, which causes global
warming. Remember last year, when it hardly snowed all winter and you couldn’t use your new
sled?”
“ What’s ethanol?”
“ Ethanol is a fuel made from plants like corn that take carbon dioxide out of the air, so when we
use ethanol instead of gas we don’t hurt the earth’s climate.”
Mrs. Greensleeves proceeded to fill her flex- fuel car with ethanol distilled from corn in a distant,
coal- fired plant, and shipped a long way. The corn came from farms that had recently intensified
their production in response to the growing demand for biofuels, switching from alternating corn
with soybeans to every- year corn and applying more fertilizer to increase yields. Mrs.
Greensleeves would be surprised to learn that the ethanol she pumped that day had essentially
the same global warming impact as the gasoline she had so thoughtfully avoided. Her good
intentions were thwarted by a broad policy failure that this study addresses.
Because biofuels can be evaluated for their real contribution to environmental goals, government
could help Mrs. Greensleeves by requiring environmental performance labels, or even by
providing tax credits for environmentally preferable ethanol production. Then Mrs. Greensleeves
might see lower prices at the gas stations that used ( for example) ethanol made in biomass-powered
factories. Even her indifferent neighbor, Mr. Brown, would then have an incentive to
fuel his car in an environmentally responsible way. And other states could set their own
standards according to local interests, so that ( for example) Mrs. Greensleeves’ brother- in- law in
the next state over would be assured that all of the ethanol available where he lived was good for
the environment, because fuel with poor environmental performance had been banned entirely.
Which policy approach is best depends on many political and contextual factors, but both depend
on being able to distinguish environmentally friendly biofuels from environmentally inferior
types. In this study we show that it is possible to make such distinctions and offer some practical
suggestions about how to do so, and thus create markets for green biofuels.
1 Introduction
Markets for biofuels— liquid transportation fuels from biomass that replace petroleum- based
fuels— are growing rapidly around the globe, driven by government regulation and subsidies as
well as high petroleum prices. Support for these government policies has three sources: a desire
to support agriculture, to reduce the use of imported petroleum, and to improve environmental
quality ( especially global warming due to carbon dioxide, CO2, emissions from fossil fuels).
However, the environmental impacts of biofuel production and use are not measured.
This study focuses on the third rationale for biofuels, environmental improvement. It is
motivated by our recognition that treating all biofuels as generally “ green,” whether on the part
of consumers or policymakers, is wrong because of large differences in the environmental
benefits not only of different biofuels, but of the same fuel made in different ways. At present,
neither government nor consumers have any way of knowing whether using any particular
biofuel is good, indifferent, or bad for the environment. When biofuel markets were relatively
small this was deemed acceptable, in part to allow the industry to develop. However, as the
industry has entered an explosive growth phase, it is well past time to address the environmental
performance of biofuels.
To support markets for green biofuels, a Green Biofuels Index is needed to provide a framework
for measuring and communicating the environmental performance of biofuels. This performance
can be communicated in many ways, from consumer information to producer incentives to
regulation. We discuss the key concepts involved in creating such an index, show three ways an
index could be constructed, examine seven possible implementations, and provide examples of
how biofuels might perform on these indices.
Ignoring the differential environmental effects of particular biofuels made in particular ways is
unwise, for several reasons. First, the biofuel industry is supplying nearly 5 percent of the total
U. S. gasoline market, growing rapidly, and very profitable. Government policies to further
subsidize, mandate, and otherwise promote biofuels are being implemented, and more are
proposed. Given the large investments in research and capital that continue to flow into the
biofuels sector, it is time to provide incentives and requirements for high environmental
performance so that the economy is not saddled with the legacy costs of shortsighted
investments. Second, biofuels are now being proposed as solutions to environmental problems,
especially climate change, and good management of any issue requires high- quality information
about achieving appropriate goals. Third, new feedstocks and new processing technologies are
now emerging, with many more in the wings, so providing appropriate incentives for the
commercialization of the cleaner of these approaches is critical.
Current government policies tend to ignore both the environmental costs and the environmental
benefits of biofuels. The present market will not achieve a socially optimal outcome because
these effects are neither captured in biofuel prices nor visible to decision makers. This study
outlines the major positive and negative environmental consequences of different biofuels’
production and explains how we can maximize the benefits of biofuels through the measurement
and management of their environmental consequences. We propose a green biofuels index in
order to provide a systematic way to communicate the environmental merits of any given
Creating Markets For Green Biofuels 2
biofuel, to ( a) enable a market for clean “ green biofuels,” ( b) promote innovation in the biofuels
sector, ( c) provide consumers with information about the environmental attributes of different
fuels, and ( d) establish a basis for regulatory action. The index includes quantitative measures of
greenhouse gas emissions and petroleum consumption for each fuel pathway, and qualitative
measures of the environmental effects of feedstock production.
Maximizing the overall social benefits of biofuels therefore requires a reliable index of
environmental performance, or a green biofuels index. Creating such an index would enable a
wide array of possible implementations, seven of which are discussed in Section 5.
As an example, we illustrate how a green ethanol index could be constructed for domestically
produced ethanol. Other biofuels ( such as biodiesel) are important, but we have chosen to focus
on ethanol because it is currently, and will be for the near future, the largest existing biofuel in
production and use in the United States, the beneficiary of large subsidies and regulatory support
( such as fuel content mandates), and the subject of active debate as to its environmental benefits.
The themes and methods discussed here for certifying the “ greenness” of U. S. ethanol
production are largely applicable to other fuels ( like biodiesel) and production contexts ( such as
tropical palm oil production), with some significant additional variables introduced by alternative
technologies, unique concerns of other locales, and the exigencies of international trade. An
important gap that must be filled by future research is how a green biofuels metric could work in
an international context, for both energy and agricultural markets are global. In addition, this
study is limited to environmental performance and does not consider important social and
economic issues, which should be addressed in future efforts to develop sustainable energy
systems.
U. S. agriculture is clearly capable of producing biofuels with high environmental performance,
and many existing producers could achieve very high scores under the indices we propose, if
they are given incentive to do so. Using a green biofuels index in one of the implementations
identified above would allow producers to differentiate their products and command higher
prices by using environmentally superior practices. Consumers would know when their
consumption of biofuels was improving environmental quality, and government could have a
basis for supporting biofuels that improve public value. Thus, a market for green biofuels might
begin to develop, one that could eventually replace some of the current agricultural subsidy
system and lead to a stronger agricultural economy in the United States.
The environmental benefits of biofuels must be evaluated based on the “ lifecycle” of production,
use, and disposal. 1 For instance, corn, the primary ethanol feedstock in the United States, is often
responsible for significant environmental harm, including damage to water quality, soil, and
biodiversity. And converting the corn feedstock to liquid fuels can, depending on the technology
and energy sources used, consume large amounts of water and result in significant greenhouse
gas and other air emissions. Looking at tailpipe emissions alone is not enough.
1 The concept of “ lifecycle” can also include supply infrastructure and end- use equipment. In this study, we assume
that biofuels are liquids that can be blended into, or substitute for, liquid hydrocarbon fuels relatively easily. In this
case, the changes to supply infrastructure and end- use technologies necessary for the use of green biofuels will be
small ( especially relative to biofuels with poor environmental performance); therefore, we ignore them.
Creating Markets For Green Biofuels 3
For instance, increasing corn production could increase soil erosion and nutrient runoff and even
push agriculture into natural habitat land. And when ethanol plants burn coal for power, the
resulting biofuel can be essentially equivalent to gasoline in terms of lifecycle greenhouse gas
emissions. On the other hand, biofuels production can also have positive impacts on the
environment. Converting row crops to perennial crops such as switchgrass, for example, reduces
erosion, water consumption and chemical use while significantly increasing soil carbon.
The wide variety of biofuel feedstocks, processing technologies, coproducts, and fuel
formulations makes biofuel policy complex. For example, biodiesel can be made from a range of
feedstocks, including soybeans, canola oil, palm oil, and restaurant waste oil. Ethanol can be
made from food crops such as corn and sugarcane; from numerous “ cellulosic” feedstocks
including purpose- grown poplar, willow, and switchgrass; or from agricultural residues, timber
industry waste, and municipal solid waste— all with different environmental impacts. In addition,
biofuel production facilities can use a range of energy sources for heat and power ( e. g., natural
gas, coal, wood chips, corn stover, and manure), resulting in drastically different greenhouse gas
emission profiles.
This study does not focus on net energy or on reductions in petroleum consumption, because
neither helps decision makers. Net energy is not a useful metric because it combines different
types of energy that have very different uses and values and should therefore not simply be
added together ( Farrell, Plevin, et al. 2006). Petroleum scarcity is not as much an environmental
issue as are the consequences of petroleum use and production, especially the use of low- quality
petroleum resources. ( Farrell and Brandt 2006). Moreover, while the “ energy security” benefits
are a driving political force behind biofuel policies, lifecycle assessments ( LCAs) consistently
show that ethanol and biodiesel production uses very little petroleum, regardless of the
production pathway ( Sheehan, Camobreco, et al. 1998; Wang 2001; Farrell, Plevin, et al. 2006). 2
Substituting any biofuels for liquid fossil fuels reduces petroleum consumption, so there is little
value in discriminating among them on this basis.
Of course, ordinary fossil- based fuels are not green either, and biofuels should be compared on
an equal basis to the entire range of available transportation fuels to allow for a fair comparison
and choice among all fuels. A green biofuels standard could burden biofuels with stricter
standards than we apply to conventional fuels, which would be inappropriate. The preferred
approach would be to develop a green fuel standard applicable to all fuels, perhaps following the
model in California, where a fuel- neutral Low Carbon Fuel Standard is being developed
( Schwarzenegger 2007). However, this task is beyond the scope of the current study and must be
left for future research.
With these limitations, the recommendations in this study can still be applied to a wide array of
biofuels in different places around the world. The most general description of the approach we
recommend is to measure important performance characteristics with rigorous methods so that a
variety of policy measures can give fuel producers and consumers incentives to improve the
performance of these fuels.
2 Note that some researchers obscure this fact by using petroleum- based energy units to measure the fossil fuel
inputs to biofuel production, even though the vast majority of fossil fuel inputs to biofuel production today are coal
and natural gas, not petroleum.
Creating Markets For Green Biofuels 4
2 How Biofuels Are Produced
Biofuels are produced in two distinct stages, feedstock production ( or collection) and processing
( sometimes called conversion or biorefining). Figure 1 shows the place of biofuel production in
the larger agricultural production system, and shows the major inputs and environmental
concerns with each stage. On the left is the feedstock phase, which is illustrated as crop
production. In the center is processing, represented as a biorefinery. This study considers these
two phases.
On the right are some of the important markets into which biofuels are sold. Note that biofuel
production generally yields one or more coproducts, or may be a coproduct of some other,
higher- valued product. ( A type of animal feed is often the coproduct of corn ethanol, while
biodiesel is often thought of as a coproduct of soymeal.) Many of these markets are global.
Figure 1 illustrates the crucial idea that biofuel production affects many different markets,
including markets for inputs ( e. g., land and water) as well as markets for agricultural products
and biofuel coproducts ( e. g., food and animal feed). For this reason, many factors can affect the
costs of producing biofuels and the prices at which they can be sold. Similarly, many factors can
affect the environmental effects of biofuel production.
Figure 1: General Biofuel Pathway with Inputs and Environmental Impacts ( simplified)
Note that in addition to causing environmental effects, such as soil erosion and GHG emissions,
biofuel production and use also displaces some environmental effects because they substitute in
FEEDSTOCK
Crop
Production
( farming)
Capital
( equipment)
Land
Energy
Fertilizers
Water
Pesticides
Environmental Effects
- Soil Erosion
- Nutrient Runoff
- Pesticide Runoff
- Land Use ( Conversion)
- GHG Emissions
- Biodiversity
PROCESSING
Biofuel
Production
Capital
( equipment)
Energy
Water
Environmental Effects
- Air Pollution
- Water Pollution
- GHG Emissions
MARKETS
Food
Export
Animal Feed
Other ( chemicals,
electricity, etc.)
Fuel
Inputs
Inputs
Environmental Effects
- Displaced Nutrient and
Pesticide Runoff
- Displaced Air Pollution
- Displaced Water Pollution
- Displaced Land Use
- Displaced GHG Emissions
Creating Markets For Green Biofuels 5
fuel and other markets for products that have their own environmental effects. At small
quantities, displacement may be a reasonable way to think about these effects, but as biofuels
grow in magnitude, it will become more important to identify and quantify how biofuels affect
markets by effectively increasing supply and therefore the amount demanded.
These interactions vary greatly by fuel and pathway ( as illustrated in Figure 2), so any attempt to
illustrate a comprehensive set of biofuel pathways and related markets would quickly become
overwhelming. This is especially true because different production pathways will often involve
competition and substitution among inputs and coproducts. In most biofuel production today, the
animal feed market is among the most important because it is large and because most agricultural
biofuel production yields a fraction of low- quality product that is salable only as animal feed.
Just two biofuels are currently in commercial production: ethanol and biodiesel. In the future,
additional options may become available, including bio- butanol and biomass- based Fischer-
Tropsch diesel. Only current biofuel feedstocks and conversion pathways are discussed in this
section.
2.1 Feedstock production
Most biofuel feedstocks are presently produced through conventional agricultural activity. Major
commodity crops are raised in large- scale, highly industrialized agricultural operations. The vast
majority of biofuel consumed in the United States today is domestically produced corn ethanol. 3
As of 2005, domestic ethanol production was about 40 times larger than domestic biodiesel
production, on an energy basis. 4 The primary biodiesel feedstock used in the United States is
soybeans, the second- largest crop grown in the United States after corn. 5 ( See the Corn
Production box, below.)
In the medium- term future, ethanol feedstocks will include lignocellulosic materials such as
agricultural and forestry residues, timber industry and municipal waste, manure, and energy
crops such trees and grasses. Unlike the production of energy crops, the utilization of residues
and wastes requires no additional land use, and thus biofuels produced from these feedstocks do
not compete with food or fodder. It is important to recognize, however, that agricultural and
some forest residues ( for example, slash)— often miscategorized as waste— serve agronomic and
environmental purposes such as reducing soil erosion, providing wildlife habitat, and improving
soil quality. Thus, the quantity safely, or wisely, available is limited.
3 Ninety- five percent of U. S. ethanol is made from corn, 70 percent of which is produced in the top four corn-producing
states of Iowa, Minnesota, Nebraska, and Illinois. Ethanol is also produced from wheat, sorghum, and
brewery or dairy waste, but these sources constitute a small percentage of the market. Similarly, biodiesel is
produced largely from soybeans, but it is also being domestically produced in relatively small quantities from
canola and sunflower oils and restaurant waste oils and grease. Given their overwhelming dominance as
feedstocks, we focus our discussion on corn and soybeans.
4 2005 ethanol production was approximately 4 billion gallons, whereas biodiesel production was 75 million gallons.
See http:// www. biodiesel. org/ pdf_ files/ fuelfactsheets/ production_ graph_ slide. pdf, www. ethanolrfa. org.
5 Throughout the 1990s, 95 percent of soybean acreage was in rotation with other crops, predominantly corn. See
Padgitt, et al., “ Production Practices for Major Crops in U. S. Agriculture 1990- 1997.”
Creating Markets For Green Biofuels 6
Box: Corn production
Corn is the largest crop in the United States by acreage ( U. S. Dept. of Agriculture 2005) and
receives the most fertilizer and pesticide per unit area of any major crop ( Padgitt, Newton, et al.
2000). These nutrients and chemicals have had detrimental effects on groundwater and surface
waters, especially in the Corn Belt and downstream through the Mississippi River to the Gulf of
Mexico ( Battaglin, Furlong, et al. 2001; Capel, Hamilton, et al. 2004). Most farms in corn
production use some form of conservation or reduced tillage; only about a third use conventional
tillage, which has higher erosion rates. Many farmers practice rotation of crops, often growing
corn in annual or 2/ 1 rotation with soybeans. This practice reduces the external nitrogen fertilizer
needs of the corn crop and disrupts pest lifecycles, reducing pesticide needs ( though the practical
difference for farmers’ practices may be smaller than advocates had expected).
One reliable study of potential domestic bioenergy production from agriculture and forestry and
some of the cellulosic content of municipal solid waste ( MSW) found that as much as 1.3 billion
tons of cellulosic feedstocks may be technically available annually ( Perlack, Wright, et al. 2005).
This feedstock could theoretically produce enough biofuel to replace one- third of current gas
consumption. While no commercial- scale cellulosic ethanol facilities are currently operating,
several demonstration plants are in operation in the United States, Canada, and Europe, and
several commercial- scale facilities are now planned.
2.2 Fuel processing
Biofuels production facilities increasingly deserve to be called biorefineries. Rather than
producing only biofuels, a biorefinery can convert one or more feedstocks into a range of
products, including biofuels, electricity, animal feed, and, eventually, other value- added
chemicals. Conversion processes differ by feedstock and the slate of coproducts desired. We first
discuss ethanol production, and then biodiesel production.
Ethanol
Fuel ethanol is produced much as alcoholic beverages have been for millennia: Yeasts are used
to ferment sugars into alcohol, which is separated from water by distillation. Differences in fuel
ethanol production processes are largely based on what is required to make sugars available to
the yeast.
For “ sugar crops” like sugarcane, sweet sorghum, or sugar beets with high native levels of
sucrose, all that is needed is to press or soak out the sugar syrup. Crops like corn, wheat, or grain
sorghum are made up mainly of starch, which is a chain of many sugars connected together.
Producing ethanol from these crops first requires converting the starch to sugars ( sucrose and
glucose) in a process called saccharification. In practice, saccharification is accomplished by
grinding the starch- containing grains, adding water to create a slurry, and then adding enzymes
that break down the starch to sugars. Finally, in the case of cellulosic ethanol, woody or
herbaceous biomass first must be subjected to relatively intense treatment with heat, acid, or
additional enzymes to make the complex carbohydrates in cellulose available for
saccharification. This step has been the main obstacle to economic cellulosic ethanol.
Importantly, the final product of all these pathways is exactly the same: Ethanol is a simple
molecule, and there is no way to distinguish finished cellulosic ethanol from corn ethanol.
Creating Markets For Green Biofuels 7
The result of any of these initial processes is sugar syrup. Yeast is added to ferment the sugar to
alcohol, which is distilled several times to increase the alcohol strength to 95.6 percent and
finally forced through a molecular sieve to achieve 99.5 percent ethanol. The last step in
producing fuel- grade ethanol is the addition of a small amount of a “ denaturant” to render the
alcohol undrinkable and thus exempt from beverage alcohol regulation. Most fuel ethanol in the
United States contains 5 percent gasoline as the denaturant.
Today’s ethanol plants include older and newer facilities, almost all of which use corn kernels as
their feedstock. Early ethanol plants were food- processing facilities in which ethanol production
was merely one of many processes, rather than the primary focus of plant design and operation.
Thus, these older plants use “ wet milling,” a process that allows the simultaneous production of
several commodities from whole corn, including corn oil, corn gluten, and germ meal. From the
corn starch, either high- fructose corn syrup or ethanol can be made.
In contrast, almost all new corn ethanol plants, and now the majority of plants in production, use
the “ dry grind” process, a simpler and more efficient way to produce ethanol but not the other
commercial products of corn. Dry- grind plants ferment the whole crushed corn kernel and
separate out its one coproduct, distillers grains, 6 from the solids left after fermentation. Most
distillers grains ( DG) are used as animal feed for dairy cows, beef cattle, swine, and poultry.
Most ethanol facilities process corn grown within 30 to 40 miles. This minimizes transportation
costs and is also a reflection of the local, cooperative ownership of many facilities.
Approximately one- third of domestic ethanol production capacity is cooperatively owned.
Biodiesel
Biodiesel is typically produced in a two- step process in which oils are first extracted from lipid-bearing
biomass feedstocks ( in the United States, most often soybeans) and then converted to
fuel. Extraction involves crushing the oilseed and using a chemical solvent ( often hexane) to
extract the oil. The resulting oils are reacted with an alcohol ( typically methanol) in the presence
of a catalyst to produce methyl esters ( biodiesel) and glycerol as a coproduct. Crushing soybeans
also yields soymeal, which is a valuable animal feed.
The market for biodiesel in the United States today is quite different from the U. S. ethanol
market. Although biodiesel is typically compatible with existing diesel engines without
modification, oilseed crops, in the United States at least, have comparatively low yields of fuel
per acre ( 50– 100 gallons per acre for soy biodiesel vs. 300– 500 gallons per acre for corn
ethanol). Further, biodiesel is essentially a coproduct to soymeal. Until and unless these basic
facts change, domestic biodiesel is likely to remain expensive and its market small.
Figure 2 illustrates the place of corn ethanol and soy biodiesel among the many markets in which
they participate, and shows just how complex the interactions of these two can be. The largest
market for both corn and soybeans is domestic animal feed, which accounts for more than half of
all U. S. corn and soybean consumption. Exports ( for both food and animal feed) are the second-
6 Distillers grains may be mixed with the condensed solutes from fermentation and may be sold wet or dry. The most
common formulation is dried distillers grains with solubles, hence the common abbreviation DDGS. Here we use
the more generally applicable DG.
Creating Markets For Green Biofuels 8
biggest market at about 18 percent of consumption. Ethanol consumed about 13 percent of U. S.
corn in 2004 and has begun to take market share away from exports. Food and other uses account
for about 13 percent of U. S. corn production. Soy biodiesel, as a coproduct of animal feed, is not
counted as a separate component of soybean production.
Dry- grind ethanol plants sell into two markets, gasoline and animal feed, while biodiesel
production has three product markets, diesel, glycerol, and animal feed. One important feature
not show in this figure is how corn and soy compete for land. Most corn is grown in rotation
with soy, with one soy crop between one to three corn crops, largely because soy is a legume
whose roots host bacterial colonies that add nitrogen to the soil beyond what the plant requires.
Therefore, while corn production can be increased by going to all- corn rotations, this lowers per-crop
yield, requires more chemical fertilizer, and causes additional soil erosion than corn- soy
rotations.
Figure 2: Key U. S. Energy Crop Production Pathways ( simplified)
2.3 Environmental consequences
The two stages of biofuel production, feedstock and processing, pose fundamentally different
challenges in measuring environmental performance. Feedstock production is highly diverse, is
linked to many other processes, and has some effects that are not only highly variable because of
weather and local conditions, but also very difficult to measure. Therefore, a green biofuels
metric will have to rely on a combination of measured, modeling- based, and practice- based
methods for evaluating environmental performance for feedstock production. In contrast,
processing has much more limited and measurable environmental effects. Each is discussed in
turn.
FEEDSTOCK
Corn
PROCESSING
Dry- Grind
Ethanol Plant
MARKETS
Domestic Food
Export
Animal Feed
Glycerol
Soy Diesel
Soy Crusher &
Biodiesel Plant
Gasoline
DDGS
Soy Meal
Creating Markets For Green Biofuels 9
2.3.1 Feedstock
A key analytic requirement for an environmental index is that any environmental harm caused by
raising a feedstock crop be attributed to the fuel and its coproducts. Thus, addressing the
environmental effects of agricultural production is a necessary component of an environmental
index for biofuels. Feedstock production entails the same variety of environmental risks or
damage–– which depend as much on farming practice as on the crop in question–– as those that
result from any type of agriculture.
Most of the environmental impacts from feedstock production occur on the farm, in growing and
harvesting crops or in removing crop residues. Greenhouse gases are released by burning fossil
fuels in most farm operations, and microbial activity in the soil releases significant quantities of
nitrous oxide ( N2O), a powerful greenhouse gas, primarily as a result of fertilizer application.
Soil quality suffers as tillage and cultivation expose soil to wind and water erosion, hastening
soil loss from land and siltation in rivers. The use of heavy machinery compacts the soil,
reducing water and oxygen availability, resulting in declining soil quality. Water used in
irrigation may be an environmental concern if it is pumped from overdrawn aquifers or
transported from distant basins. The infiltration or runoff of excess nutrients results in
groundwater contamination as well as algal blooms and oxygen- starved water in aquatic
ecosystems downstream. The use of pesticides causes pollution of surface water and
groundwater and unintended harm to humans and wildlife. Removing crop residues increases
soil erosion, reduces soil organic content, removes nutrients, lowers yield, and consumes fossil
fuels ( Wilhelm, Johnson, et al. 2004). For more detail, see Appendix A: Measuring Multiple
Dimensions of Environmental Performance.
The use of crops for biomass feedstock can affect markets for global commodities like corn,
because the crops use land that would otherwise have other uses; the production of biofuel
feedstocks displaces these other uses. This displacement induces economic effects, which in turn
can induce changes in land use elsewhere. These effects are not well characterized today and are
excluded from current analytical methods, suggesting an important area for further research
( Delucchi 2004).
For instance, increased demand for corn by ethanol plants in the United States appears to have
reduced U. S. corn exports and raised the price of corn on global markets. Such a price increase
will both reduce demand ( with possible consumer welfare impacts) and create incentives for
more land in the exporting nation ( the United States) and importing nations to be put into corn
production ( extensification). Thus, the use of U. S.- grown corn for ethanol production can induce
the conversion of previously uncultivated lands elsewhere, a phenomenon called “ leakage.” ( The
effects of an action, growing corn for ethanol production, has leaked, in this case from where it
might be controlled by a green biofuels program to where it will not be.) A green biofuels system
must recognize this potential for leakage of environmental impacts to systems outside the strict
biofuel system boundary, however how to do so is not clear and should be the subject of further
research.
2.3.2 Processing
The environmental consequences of biorefining are few, and they are easily quantified and
managed. They include energy use, which results in emissions of volatile organics, toxics, and
Creating Markets For Green Biofuels 10
greenhouse gases; and water use in processing and in boiler system cooling. The specific impacts
vary with feedstock and energy source.
Biofuel production typically requires both thermal and electrical energy. Ethanol producers today
use a variety of fuel sources ( e. g., coal, natural gas, biomass) and energy conversion technologies
( combustion, gasification, cogeneration) resulting in a range of environmental outcomes.
Typical dry- grind corn ethanol facilities burn natural gas for heat and buy electricity from the
grid. However, in response to higher natural gas prices, several U. S. dry- grind plants are
exploring or deploying innovative alternatives to natural gas. Some plants are being developed or
redesigned to use coal, and others are gasifying or combusting wood waste, distillers grains, and
corn stover or using advanced cogeneration units ( Nilles 2006). Others are locating near cattle
feedlots to sell wet distillers grains, halving a typical plant’s natural gas consumption by not
drying the coproduced distillers grains. The challenge for policy makers is to ensure that
incentive structures encourage the more socially beneficial configurations and energy sources,
and that they discourage much less beneficial options such as switching to coal. The GHG
profiles of several biorefineries are detailed in Section 6.
The environmental impacts of the dominant corn ethanol dry- grind process include water
consumption, air emissions from fuel consumption and drying distillers grains, and carbon
dioxide from both fermentation and fuel consumption.
Water consumption is a particular concern in the Midwest where competition for water supplies
is increasing ( Keeney and Muller 2006). According to a report by the Institute of Agriculture and
Trade Policy ( IATP), only one state, Minnesota, tracks water consumption by ethanol plants. The
average water consumption rate in Minnesota declined from 5.8 gallons of water per gallon of
ethanol in 1998 to 4.2 gallons in 2005, with most plants using 3.5 to 6.0 gallons ( Keeney and
Muller 2006). New plants reportedly use 3 gallons of water per gallon of ethanol. IATP estimates
the average in 2006 was 4.0 gallons.
Because the production of biodiesel is much simpler than the production of ethanol, the
environmental implications are fewer. They include water consumption and greenhouse gas
emissions from fossil fuel combustion, and hexane volatilization.
Creating Markets For Green Biofuels 11
3 Measuring and Communicating Environmental Performance
To improve the environmental performance of biofuels, so that they really can be called “ green,”
requires appropriately measuring environmental performance and communicating the results
with information and incentives. However, the best way to do so varies with the type of
environmental impact under consideration and the purpose of the communication. Describing the
environmental consequences of the two biofuel production stages— feedstock production and
processing— may require different tactics. The distinct nature of each phase, the goals of the
regulatory program, and the state of the art for various tools will determine the best approach.
For reasons outlined below, directly measuring the environmental performance of agriculture
will likely remain infeasible. Agro- environmental models may eventually allow accurate
estimation of the environmental performance of individual producers, but in the short term,
feedstocks can be characterized only by using approximate or categorical performance measures,
even for environmental effects that are in principle quantitative ( e. g., GHG emissions). In
contrast, the specific environmental consequences of processing can typically be measured
quantitatively at a reasonable cost.
A number of existing certification systems may be relevant, including the USDA Organic, Forest
Stewardship Council, U. S. Green Building Council’s LEED ® sustainability standard, Green
Gold, and the United Kingdom Renewable Transportation Fuel Obligation. These programs are
discussed in Appendix B.
There are three general approaches to measuring environmental performance: direct observation,
which is quantitative; modeling, which is also quantitative; and indirect, qualitative
categorization. The following section describes the use of these in relation to the stages and
impacts of feedstock production.
It is important first to identify the distinction between average and marginal cases. Unlike
aggregate lifecycle analysis, which is used to assess the lifecycle impacts of an average unit of a
good, a green biofuels index would be used to measure and communicate the specific, individual
impact of each unit of biofuel ( or at least the average impacts of an individual batch of fuel). The
discussion that follows should be understood as a brief examination of an environmental index
that could be constructed for a unit of a particular fuel ( for example, “ ethanol produced between
June 1 and June 7 in the Smith Refinery”).
3.1 Feedstock
Agriculture is a complex, semi- natural system, but prevailing approaches to agriculture are
widely criticized on environmental grounds. Changes to prevailing practices can have an
extensive mix of intended and unintended effects, both directly on the farm and on upstream and
downstream processes ( Kulshreshtha, Junkins, et al. 2000). For example, reducing nitrogen
application may reduce both the nutrient runoff and N2O emissions from soil, but it also
decreases yield, which can cause more land to be converted from a natural state into production
elsewhere. One way to reduce nitrogen runoff is to use “ precision” fertilization methods, but
these can entail more frequent passes through the field with tractors and equipment, and so
greater diesel fuel consumption on the farm. Similarly, reducing the use of herbicides often
Creating Markets For Green Biofuels 12
requires increased mechanical or flame- based weed control, which can increase soil disturbance
and erosion, fossil fuel use, and GHG emissions.
Measuring, or even estimating, the exact environmental impacts from specific agricultural
production systems is particularly vexing. Agriculture is a classic “ nonpoint source” of emissions
that occur over an entire landscape, without a convenient smokestack or drainpipe at which to
measure them. While researchers have created experimental systems to measure emissions on
small plots, there is no practical way to directly measure soil erosion, nutrient runoff, or pesticide
drift on actual fields, especially the millions of acres of U. S. agriculture.
Not only are impacts difficult to measure directly, but their complexity and site- specificity means
that estimating or modeling emissions is difficult. For instance, agricultural soils emit N2O, a
powerful greenhouse gas, roughly in proportion to the rate of nitrogen fertilizer application or
atmospheric nitrogen fixation. Actual emissions, however, depend on several site- specific factors
including agronomic practices, temperature, and moisture. Moreover, the emissions are highly
variable, both spatially and temporally. Thus, N2O emissions can vary widely across a single
field, even over distances as short as several inches, and emission rates can vary by orders of
magnitude over the course of a year ( Skiba and Smith 2000; Gibbons, Ramsden, et al. 2006).
Finally, nitrogen leached from an agricultural field may later result in N2O emissions from the
aquatic systems to which it flows.
For these reasons, determining the environmental impacts of a unit of fuel from its agriculture
phase requires either accurate modeling or practice- based indices.
Biofuel policy would be tremendously strengthened by the use of accurate, robust, and
manageable agro- environmental models for estimating the actual environmental performance of
biofuel feedstock production for use in regulation. Agro- ecosystem models offer an alternative to
measurement in quantifying the environmental performance of agriculture. These models use
site- specific data on soil, climate, and practices to predict associated impacts, including erosion,
soil organic content, nutrient and chemical runoff, and greenhouse gas emissions. Ideally, agro-ecological
models would require a relatively manageable set of input data yet allow for the
characterization of environmental impacts from a specific set of fields, crops, and practices. In
short, modeling would allow the quantitative measurement of agro- environmental performance.
An ideal agro- environmental model for biofuels assessment would allow farm managers to
customize the baseline conditions of their farm, using historical climatic frequency distributions
and soil type distribution and average slopes for a finite number of field units. On this foundation
would be modeled the specific crop in each year, with the field operations performed and inputs
added. Finally, a small set of tests– such as crop tissue nutrient tests– might be performed to
gather additional data. The resulting calculations would yield quantification of nutrient runoff
and leaching, pesticide runoff, soil erosion, and GHG emissions. Because this model could be
installed on the farmer’s home computer, she could use it to estimate the effect of changes in
practices on performance indicators of interest, and so calculate the tradeoffs of changes in
practices versus changes in performance— and then compare the respective costs and benefits of
each.
Creating Markets For Green Biofuels 13
In this way, accurate models that allow for the quantification of agricultural performance would
not only create more powerful and useful regulation, but would strengthen the feedback and
learning process for farmers. Models that allow farmers to predict the relationship between
practice and performance would lead to better choices among current practices, and would
support cost- efficient innovation as farmers devise new solutions and methods. Also, as
modeling can be used in performance standards, it allows a greater diversity of performance
regulations to be used. Performance- based standards are likely to result in more cost- effective
improvements in environmental outcomes than are practice- based standards.
However, the state of the art in agro- environmental modeling is inadequate for the purposes
discussed here. For one thing, some researchers have questioned the accuracy of existing models
in predicting specific emissions from specific fields ( Cassman 2006; Baker, Ochsner, et al.
2007). Moreover, models of multiple environmental impacts are not well integrated. Finally, the
administrative burden to farmers or regulators of implementing many current modeling
approaches could be high. Improving, integrating, and streamlining these modeling approaches
should be a particular priority of future biofuel research.
For the reasons outlined above, it does not appear possible at this stage to measure the
environmental outcomes of specific agricultural practices by specific producers. Therefore, less
specific methods must be used.
Instead of specific emissions or accurate emissions models, what can be observed are the relative
performances of different categories of crop, farm conditions, and farmer practices. To the extent
that certain sets of crops, conditions, and practices consistently result in superior results, they can
be identified as “ best practices.” These best practices, tailored for each farm, can be reliably
predicted to reduce negative environmental impacts. For instance, between annual and perennial
crops, all else being equal, the perennial crop will exhibit lower erosion, nutrient runoff, and
greenhouse gas emissions. Between corn grown in non- irrigated Minnesota and irrigated
Nebraska, Nebraska corn will have higher water use and greenhouse gas emissions. And between
corn grown with conventional tillage and corn grown with conservation tillage, all else being
equal, conservation tillage will result in lower emissions. Some of these characteristics,
specifically crop and region, are distinct and robust enough to be used directly in a qualitative
assessment. The majority of best practices, however, must be determined as a set, in the context
of the whole farm.
The use of this comprehensive best- practices approach consists of three steps:
! " a resource assessment, detailing the unique characteristics of the farm and surrounding
environment including soil type, climate, water availability and quality, terrestrial and
aquatic habitat, and vulnerability to the impacts of farming, including soil erosion, chemical
runoff and drift, and greenhouse gas emissions;
! " resource management plans that propose mitigation measures to reduce each potential
negative impact below a specified threshold. Indicators may be used to identify this
threshold; and
! " assurance, review, and adaptation programs ensuring that the management plans are carried
out, that their effectiveness is periodically assessed, and that adaptive management occurs to
revise and refine plans where necessary.
Creating Markets For Green Biofuels 14
The primary goals for best practices should be to reduce water depletion ( i. e., usage in excess of
recharge rates), soil erosion, agrichemical runoff and drift, and GHG emissions. In addition,
feedstock producers should avoid environmentally harmful land- use change involving habitat
destruction, deforestation, or the conversion of grasslands to row crops. They should also
eliminate the use of the most toxic pesticides.
A good model of such a resource management best- practices program is found in the USDA’s
Conservation Security Program ( CSP), enacted as part of the 2002 Farm Bill. Under this
voluntary program, farmers develop and implement resource management plans specific to their
farmland in return for five to ten years of clearly defined per- acre annual payments ( McKnight
Foundation 2005). Of course, other examples exist as well.
Adaptive management of best- practices approaches to feedstock production must take a whole-system
view. That is, practices at the farm level should be reviewed and modified to support the
best outcomes possible in light of the farm’s unique circumstances, but also the palette of
practices and evaluation methods should be continually monitored, reviewed, and updated to
reflect evolution of the applicable science, advances in technology, changing environmental
priorities, and shifts in the relative costs of inputs. All this must be accomplished in an
environment of rapid growth and technological innovation in biofuels, where new feedstocks
may lead to new types of impacts before producers, regulators, or researchers have even learned
to recognize them.
Biomass from forestry systems have important similarities to and differences from agricultural
energy crops and biomass residues. Accordingly, some existing systems are able to encompass
the environmental performance of forestry biomass, while other forestry biomass sources do not
have well- developed criteria ( Rotherham 1999).
“ Short- rotation woody crops” ( SRWC), meaning plantings of willow, poplar, or other crops that
are grown in a coppice system for five to ten years with annual harvests of wood chips, are
appropriately captured in the same agriculture- oriented best- practices systems described above.
Biomass systems involving longer- life and larger timber species, such as eucalyptus plantations,
can be addressed under standards for plantation forests, such as the Forest Stewardship Council
plantation certification ( Forest Stewardship Council 2006). And forestry residues from FSC-certified
forests can carry the certification level of the forest, as biomass residue harvest would
necessarily be regulated under the forest certification.
However, the environmentally responsible use of biomass residues from conventional forests is
not well defined at present ( Richardson 2005). Forest thinning operations for forest health or fire
fuel reduction, commercial thinning operations, commercial logging operations, and the
processing of forest products all generate residues— but none of these sources have satisfactory
environmental performance or certification systems. This is an outstanding research need.
3.2 Processing
Biorefining, in contrast to agriculture, is a relatively simple, linear process in a controlled
environment, with easily measured environmental outcomes and established process alternatives.
Creating Markets For Green Biofuels 15
The primary environmental impacts, as discussed above, are air emissions of greenhouse gases,
volatile organic compounds ( VOCs), and toxins due to fuel combustion and drying of distillers
grains; water consumption; and emissions to surface waters.
These environmental impacts can be usefully observed and linked to market units of fuel by
direct observation or calculation, rather than by scoring or ranking practices as we think
necessary for agriculture. Emissions of criteria pollutants are already managed or measured with
standard control technologies. Greenhouse gas emissions are easily determined by the use of
fuels, e. g., coal, natural gas, biomass, or biogas. Water use and emissions can be measured by
plants.
3.3 Performance- vs. practice- based standards
The distinction between a performance- based index and a practice- based index is critical, in part
because this determines how goals are measured, how many goals can be encompassed, and the
breadth of the index’s possible applications. Briefly, a performance- based index is built on
information about the actual consequences of manufacture and use of a product, while a practice-based
index assures that certain methods were employed in production.
A performance- based index is preferable in tax/ subsidy and regulatory applications because no
one technology is necessarily privileged. Instead, the desired end results are specified, and
producers retain maximum flexibility in the means by which they meet performance goals. This
is always important for maximizing cost- effectiveness, but in the rapidly developing biofuels
industry, it is especially desirable not to place any unnecessary restrictions on technological
development. Finally, performance indices are also likely to be necessary for complying with the
non- discriminatory standards of international trade agreements.
However, performance- based indices can be created only for quantifiable policy goals, such as
greenhouse gas emissions or water consumption. “ Performance” assumes that the characteristic
of each fuel production pathway with regard to a policy goal is both observable and quantifiable.
The measurability requirement limits the strength of performance indices in regulating impacts
with high uncertainty, though uncertainty can be accommodated as long as the size of potential
error is small relative to the magnitude of the effect.
Practice- based standards can address a broad range of policy goals, including environmental
impacts that are unobservable, unquantifiable, or highly uncertain. Practice- based standards offer
a way to improve environmental performance, even if the exact performance isn’t known. In
other words, the performance of practices can be ordered, even if the magnitude of the
differences cannot.
Performance standards are preferable, when conditions permit, for two reasons. First,
performance is what we care about, while practices are not as closely linked to actual outcomes.
Furthermore, public policy ( e. g., a regulation, label, or graduated subsidy) based on a
performance index invites innovation in the ways in which performance goals are reached. The
performance standard for a fire extinguisher, for example, specifies that to receive a 1A rating, it
must extinguish a flaming, fully involved “ log cabin” of wood pieces of specified dimensions
Creating Markets For Green Biofuels 16
and moisture content in a certain time. 7 In contrast, a practice standard would refer to production
or manufacturing processes for the fire extinguisher: It might specify how much of what
chemical, under what charge pressure, the device must contain. However, practice standards
suppress technical innovation; if a better fire extinguisher chemical were discovered, it could not
be rated under a practice standard until the rating body established and promulgated an entirely
new standard.
Accordingly, we recommend performance standards where we can, in biorefinery application,
and practice- based standards for evaluating agricultural production of biofuels.
3.4 Quantitative GHG measurement
As discussed in Section 3.1, calculating the specific, quantitative environmental impacts from
specific agricultural fields is not currently feasible in most cases. This is particularly true of
greenhouse gas emissions from agriculture. In fact, to date, few best practices have been
identified which reliably reduce greenhouse gases across different farming practices. In
particular, the uncertainty range for N2O emissions from soil and from N emitted into waterways
is likely larger than the emissions differences between agricultural practices ( Farrell, Brandt, et
al. 2005), and the change in soil organic carbon under various tillage regimes is a matter of
active debate ( Cassman 2006).
In the face of such uncertainty, the characterization of greenhouse gases must retreat to the level
where categorization is robust. Therefore, to calculate the GHG emissions from biofuels
production, we recommend estimating the average emissions per feedstock type ( e. g., corn,
switchgrass, corn stover) with adjustments for large- scale regional differences that affect energy
use, such as whether the crops in the region are predominantly rain- fed or irrigated. These
regional feedstock emissions would be added to the specific biorefinery emissions to calculate
the GHG emissions for the resulting fuel.
In contrast, the measurement of GHG emissions from biorefineries is relatively straightforward:
Biorefining is a linear engineered process with clearly defined relationships between inputs and
emissions. Reliable estimates of the GHG emissions from this process can be made from a few
easily measured parameters ( generally measured per gallon of fuel produced): thermal energy,
thermal energy source ( i. e. coal, natural gas, corn stover, etc.), electricity, biofuel yield, and
coproduct yield.
3.5 Leakage
A comprehensive biofuels rating system should have some mechanism for accounting for the
possible environmental effects that arise indirectly from feedstock production. As discussed in
Section 2.3.1, the displacement of current land uses by biofuel feedstock production can lead to
more, or more intense, land use elsewhere, potentially driving a leakage of environmental
impacts from the green biofuels production chain to other, unregulated, systems.
7 ANSI 711.
Creating Markets For Green Biofuels 17
These leakage effects of biofuels production are difficult to capture, even in the aggregate, as
they require economic- agricultural- land use modeling that can isolate the effects of biofuels
production from other economic variables. It is then even more difficult to allocate these
aggregate effects to individual biofuel producers, as would be required under a green biofuels
standard. Because the effect is aggregate, there is no direct linkage from individual producers.
Thus, the fairest option would appear to be the assignment of “ average” leakage effects from
each feedstock type. For instance, if the specific effect of increased corn ethanol production on
corn prices, and the effect of increased corn prices on land conversion, could be determined, than
an average effect of each corn ethanol producer on land conversion could, theoretically, be
assigned to each producer. Such factors are not currently available but should be a focus of
research and development in a green biofuels system.
3.6 Practical implementation
The measurement and communication of the environmental quality of different biofuels should
be practical to implement. The strength of incentives transmitted to producers and consumers is
dependent on the structure of measurement, verification, and enforcement processes. Thus, the
burden placed on feedstock producers and processors and biofuel producers and distributors,
relative to the potential gain, is a critical consideration.
The cost of measuring and verifying environmental performance will increase the cost of
production, and uncertainties about the potential for higher prices for green biofuels can create
fundamental impediments to participation by farmers, processors, and biofuel producers that
would undermine the entire market, leading to potentially unacceptable market volatility and
extreme peak prices. Thus, it is crucial to any measurement and verification process not only that
cost and regulatory burden be reasonable, but also that the process be, and be seen as, feasible.
The burden of demonstrating environmental performance falls on different actors at different
points in the supply chain. Biofuel producers must source feedstock and must plan, attain, and
demonstrate the environmental performance of their facilities; suppliers, brokers, aggregators,
and distributors must track and document appropriate data along the entire supply chain.
For farmers, the link between their actions and the price premiums they stand to gain should be
clear and direct. Producers should be able to estimate in advance the likely effect of practice
changes on environmental performance, and the relation between a change in performance to
change in value of the final product. In this regard, as from a regulatory perspective, accurate
predictive agro- environmental modeling represents the gold standard: If farmers can reliably
predict how their choices will affect productivity and environmental performance at the start of
each season, they can understand how to maximize their incomes. Thus, the creativity of
producers is engaged to find the best approach for their individual farm, potentially surpassing
the consensus best practices determined by scientists and extension agents. Creating such
modeling systems should be a priority of biofuel researchers.
However, as in the regulatory setting, such accurate, integrated modeling remains out of reach
for the near future. In its absence, practice- based policies assure the farmer of receiving clearly
defined benefits for specific actions. If the adoption of a specific suite of practices is explicitly
tied to specific product quality levels, and if these levels correspond to clear market prices, the
Creating Markets For Green Biofuels 18
producer can directly compare benefits to costs. For example, if a farmer can be reasonably
assured that the adoption of practice suite A ( with an added production cost of X) will lead to a
“ four- star” rating ( discussed later) for her crop, and if she can observe that “ four- star” crops
fetch a price premium of Y relative to her current crops, then she can compare X and Y and
readily see whether the premium ( Y) is likely to be greater than the additional cost ( X). Without
this certainty, the producer faces risk that reduces the value of any premium, decreasing supply
and increasing the cost of sustainable feedstock.
Practice- based standards require a process called “ assurance,” which includes enrollment,
verification, and enforcement. It entails the development of a management plan, and regular
verification of plan compliance ( with enforcement of established sanctions for noncompliance)
by a trustworthy third party. The process should also entail a research and adaptive management
component to improve performance over time and improve the correspondence between practice
and performance.
Enrollment, verification, and enforcement are established principles under many existing
certification frameworks. For instance, the National Organic Program ( NOP) requires consistent
methodology among the 92 independent bodies accredited to certify organic producers. Steps to
ensure this include the creation of an Organic Systems Plan, ongoing documentation of certified
practices, and annual site inspections to verify compliance. However, the NOP does not include
either a pre- or post- planning performance evaluation of the effect of practices on the
environment or health. In contrast, the Conservation Security Program ( CSP) does endeavor to
determine and refine the effect of practices on the environment through site- specific evaluations
and conservation stewardship planning, but it does not include provisions for independent
verification of practices ( unless a specific complaint is filed against a producer).
The NOP is more transparent and accessible to producers— mandatory or prohibited practices are
clear and invariant, but verification is more intrusive. The CSP is less predictable— it is not
necessarily clear which practices will qualify for eligibility, and they may evolve over time— but
verification that these practices occur is minimally intrusive.
Documenting the environmental performance of biofuel processing facilities would necessitate
auditing and documentation of plant operations. Because the number of inputs and processes is
small, the burden of recording these, especially over annual operations, is likely also to be small.
Issues of industrial disclosure can be assuaged through the development of a licensed third- party
certification industry. Such an industry also serves important technical support and educational
functions, helping regulated entities to develop the capacity to cheaply meet certification
requirements.
The cost and feasibility of tracking and chain- of- custody documentation is also an important
consideration to farmers, consolidators ( grain elevators, brokers, etc.), and biofuel producers.
These issues are discussed in Section 3.9, below.
The liquidity of markets; the availability of loans, contracts, futures, and options; and the
sophistication of aggregators, processors, and distributors are also key to reducing the burden of
measurement and tracking placed on producers. These institutions reduce risk across many
Creating Markets For Green Biofuels 19
dimensions and are especially critical in the early stages of a program. A certification program
must be designed with consideration given to the ability of new and existing institutions to adapt
to the needs of the system.
The complexity of the foregoing is important, but a variety of existing schemes for certification
of many things, from the safety of foods and medicines to the accuracy of measuring devices
( including the pump that dispenses ethanol to drivers) to the authenticity of organic food, shows
that obstacles can be overcome at reasonable cost.
3.7 Dimensionality
Choices by decision makers— whether by a consumer buying fuel, a distributor supplying
retailers, or a policy maker designing regulations— to purchase or support one product over
another is intrinsically a one- dimensional process. While the decision maker may consider a
variety of relevant qualities, including the product’s price, its quality, its ability to meet his
needs, its popularity, and perhaps even its social value such as environmental performance, the
final decision is made because one product is “ better” than another. The abstract quality of
“ betterness” is necessarily a reduction of all the relevant qualities into one dimension when the
choice is made.
In effect the decision maker, at any moment, must consider a set of performance measures that
represent the contribution of each relevant dimension of choice to the scalar ordering. Moreover,
each consumer generally has her own personal tradeoff function among different product
qualities.
A subset of these relative prices pertains to collapsing various environmental performance
dimensions into the one- dimensional measure of environmental quality she may wish to use in
the larger aggregation of qualities in her overall product preference. For example, she may think
four pounds of greenhouse gas emissions are as bad as one pound of pesticide runoff into the
Mississippi River. If all consumers had the same set of performance measures, and if these were
known, products could be scored on environmental performance accurately and unambiguously
into a single number appropriate for all users. However, neither of these conditions is true;
consumer preferences vary and are known only approximately.
There is an inherent tradeoff between reporting a multidimensional list of individual
environmental performance measures versus aggregating them into a one- dimensional score. The
former may incur high costs for consumers to assess the multiple dimensions and calculate their
own tradeoffs, and it risks high costs to society if consumers are discouraged from making any
rational calculation by the cost of processing an overabundance of information. However, relying
on a predetermined aggregation system in order to report a single summary statistic loses the
efficiency and democracy of letting consumers exercise their sovereignty. A measure of limited
dimensions ensures that almost everyone will make more or less incorrect personal choices, but
it saves us all time for other things we value. Providing more information doesn’t in any case
guarantee more correct choices.
Creating Markets For Green Biofuels 20
3.8 Compensatory vs. mandatory minimums
Aggregating multiple dimensions involves tradeoffs among impacts: Better performance in one
dimension can compensate for worse performance in another. This could occur in performance-based
standards or where multiple practices are structured as “ tiers” of performance. An
alternative to this is an aggregated metric with a minimum performance requirement in each
dimension.
As an example, the LEED ® sustainability standard designates multiple compliant practice
options in each of several sustainability dimensions. These practices are designated with
increasing levels of “ points” according to their relative sustainability. These points are then
aggregated to calculate the overall sustainability level of the project ( Silver, Gold, or Platinum),
so more points in one dimension may, to an extent, compensate for fewer points in another. This
effect is limited, however, by mandatory minimum performance in each dimension for each
level.
A sharp distinction separates a so- called lexicographic ordering from a weighted one. In a
lexicographic ordering, such as is used in a dictionary, axe appears before bad even though on
average the position of the letters in the second word is earlier in the alphabet than that of the
former; nothing can compensate for the fact that a precedes b. There is no “ averaging” of the
position of the other letters.
An example for biofuels could start by assuming that the highest ranking requires that crops be
grown under a Tier III CSP contract, that fuel processing have very low impacts, and that GHG
emissions be very low. In a non- compensatory index, no biofuels made with crops grown under a
Tier II CSP contract could ever qualify for this highest rating, even if processing had no
environmental impact at all and net GHGs were negative. This example illustrates that non-compensatory
indices avoid the task of comparing incommensurate criteria, but they are rigid
and may lead to unnecessarily strict or even deceptive results.
3.9 Tracking, trading, and banking
Tracking— by which we mean maintaining an association between feedstocks and fuels and their
environmental scores across the stages of production— has two distinct purposes. The first is to
assure the correctness of any aggregate claim made about a batch of fuel mixed together from
different sources. The second is to assure the communication of incentives from downstream
consumers ( or regulated retailers) to upstream producers. The feasibility, reliability, and expense
of tracking this behavior in the variety of situations in which it may be required are important
policy considerations.
The burden of tracking is largely related to the degree of physical control necessary and the
strength of incentives to cheat. For instance, Identity Preserved ( IP) systems developed for non-
GMO crops have proven the feasibility of systems that control specific physical quanta of
agricultural products from producers to consumers. While some researchers have asserted that
the small premium earned by non- GMO crops demonstrates that IP systems impose small costs
( Bullock and Desquilbet 2002), others use a bottom- up approach to assert that the expense of
separate storage and processing facilities, in addition to recordkeeping protocol, imposes high
Creating Markets For Green Biofuels 21
direct costs and, moreover, significant indirect market barriers ( Kalaitzandonakes, Maltsbarger,
et al. 2001).
A more flexible system averages the performance across a producer’s products in a given period.
For instance, the Forest Stewardship Council’s chain- of- custody system includes a “ Mixed”
designation, under which producers of wood products may certify their product as partially
certified according to the proportion of certified wood input that is used ( Forest Stewardship
Council 2006). Under this system, physical tracking is transformed to a statistical statement of
each producer’s average performance.
Finally, tracking of feedstocks’ and fuels’ environmental performance may be wholly abstracted
from physical quanta using a credit system. For instance, feedstock producers could generate,
along with each batch of feedstock, environmental credits that could be sold and traded
separately from the feedstock. Biofuel producers similarly would purchase these credits to create
a certain rated biofuel, and the biofuel rating credits would be generated along with batches of
biofuel. These biofuel rating credits would be purchased by fuel blenders and retailers along with
batches of biofuels to support the quantity of each level of biofuel sold.
Trading, a potentially important element of a green biofuels policy, introduces important
flexibility into the market. Trading improves economic efficiency by allowing firms with poor-performing
assets ( such as older, inefficient processing facilities) to compete in the biofuel
market by purchasing credits from very green facilities, rather than face closure or very high
retrofitting costs. The green facilities, of course, would see an additional revenue stream and
might have sufficient incentive to improve their performance even more. Over time, this
arrangement, especially in the face of tightening GHG emissions restrictions, would tend to
encourage innovation and investment in green technologies and practices without inefficiently
wasting existing investments. Private firms could also enter into long- term contracts for
especially green biofuels into the future. Such contracts could be used by the buyer to hedge
risks and by the seller to obtain construction financing, creating additional green biofuel
production. In this way, trading also encourages economically efficient investments over time
( called dynamic efficiency).
Banking is the practice of holding biofuel credits, whether traded or restricted to a single firm,
from one compliance period to the next. Banking smooths the demand for green biofuels and
capital investments ( like the combination of trading and long- term contracts described above). In
addition, banking serves as a hedge against changes in market or weather conditions, and it
creates an incentive for voluntary reductions ahead of the compliance schedule. That is, banking
allows firms to overcomply, especially in the early years, and then to hold allowances as a hedge
against greater fuel demand or poor weather in the future. This encourages innovation and
investment in the near term.
A combination of trading and banking, plus the potential for long- term contracts, provides a
flexible yet robust compliance strategy without the need to “ borrow” allowances from the future.
Existing emission control programs that have used trading and banking have been very
Creating Markets For Green Biofuels 22
successful, achieving extremely high compliance rates and low costs without the need for
borrowing ( Farrell and Lave 2004). 8
The tracking system proposed under the federal Renewable Fuel Standard ( RFS) presents a
prototype foundation for a biofuels environmental credit system ( U. S. Environmental Protection
Agency 2006). Under the RFS proposal, each batch of biofuels would be assigned a series of
unique Renewable Identification Numbers ( RIN) corresponding to each gallon of fuel. These
RINs would be separable from the fuel itself and would be bought, sold, traded, saved, and
borrowed by and among fuel producers, brokers, and blenders. The purpose of the RIN is to
allow regulated entities— fuel wholesalers— to demonstrate compliance by surrendering to the
government the number of RINs corresponding to their assigned renewable- fuel production
obligation. The RFS system even foreshadows environmental tracking in its provision for the
differential generation of additional RIN for cellulosic ethanol.
This system allows for demonstration of aggregate compliance without the need for detailed
tracking of the fate of every fuel. Instead, environmental performance is monitored only at the
point of production, and compliance is enforced in the wholesale market. An environmental
performance tracking system could be similarly structured, with credits generated for feedstock
production that are purchased by fuel producers, and credits in turn generated in fuel production
that are purchased and surrendered by fuel wholesalers.
3.10 Compatibility with other regulatory structures
Many implementations of a green biofuels index are likely to interact with other policies and so
should be designed to at least minimize conflicts, and ideally maximize compatibility, with other
regulations. One of the important considerations in examining compatibility is the breadth of
coverage of the regulated sector. For instance, standards for biofuel producers will capture
almost all such plants, but standards for many types of biomass producers ( e. g., corn farmers)
will capture only a portion of such producers.
3.11 International trade
International suppliers of biofuels or raw biomass are likely to be major participants in an
expanded biofuels industry in the United States. International trade raises three important issues
that complicate measuring the environmental performance of biofuels: difficulty in determining
impacts; restrictions on regulatory standards posed by international trading agreements; and
conflict between the long- term incentives for a foreign government overseeing compliance to
establish a solid reputation for reliability and trustworthiness, and the short- term incentive to
advantage its domestic producers.
8 The successful outcome of existing market- based environmental ( mainly air pollution) regulations discussed in
Farrell and Lave ( 2004) resulted from political and regulatory processes that created unequivocal, detailed rules
that have to be defined carefully to ensure both that the resulting market is viable and that the desired
environmental outcomes are obtained. Poor market- based regulation is possible, and in one case ( part of
California’s RECLAIM program) has been adopted, but the regulatory and legal system in the United States has
prevented such ill- designed policies from being adopted or ( in the solitary case of RECLAIM) overturned the
problematic provisions.
Creating Markets For Green Biofuels 23
The feasibility, expense, and reliability of measuring and tracking biomass and fuel production
data from developing countries may be a significant obstacle, because of poorly developed
institutions and infrastructure. Other differences such as in income levels and agricultural
practices, compared with the U. S. context considered here, probably require a different green
biofuels index. Such difficulties could inhibit the application of any green biofuels index or act
as a barrier to some producers, possibly contravening the trade principles described below.
Conversely, they could provide a powerful incentive to these producing countries to develop data
collection systems that would add value to their products. For instance, hoof- and- mouth disease
was stamped out in South America mainly so its beef could be sold fresh in the United States,
but this benefited South American consumers as well.
According to expert interpretation of the text of trade agreements and the case precedent of
adjudicatory decisions, several principles should be considered in designing a green biofuels
index ( Lancaster 2006; Rogers 2006). Measures need to be consistent with World Trade
Organization rules, be based on legitimate domestic or global objectives of the importing
country, and take into account the capabilities of developing countries. Standards should be
based on scientifically sound principles with a clear nexus to health, safety, or conservation of
exhaustible resources.
Certification standards should not explicitly or implicitly discriminate between domestic and
imported products, or among the products of different foreign countries. Methodologies for green
biofuels measurement and regulation should be developed and implemented with consultation
and cooperation of international stakeholders, including all prospective export countries, and
should incorporate as much as possible existing methods developed under international
protocols.
While this study addresses the narrow issue of U. S.- produced biofuels, any environmental index
should be designed to be as compatible as possible with potential future application to
internationally produced biofuels. Future research in this area is needed.
3.12 Legibility, convention, and implicit ceilings
Any index of performance is interpreted by users against a background of social and other
conventions that we infer from related contexts. For example, hygiene ratings of food service
establishments use A as the highest category and not D, or Q, or 13, because we are conditioned
from school, and from other systems that conform to the familiar pattern, to regard A as a top
grade. In the present context, we think it important that a green fuels index not imply a ceiling on
performance. If fuels are rated on a scale from 0 to 100, a user can reasonably infer that 100 is
the most green a fuel can be, and with technical and managerial progress, this is certainly not the
case for any existing fuel. Accordingly, a criterion for any rating system is not only that it not
have a top score, but that it not appear to have one. The two examples that follow include one
that observes this discipline and another that does not, for illustrative purposes.
Creating Markets For Green Biofuels 24
4 Examples of green biofuels indices
Among the many ways to create a environmental performance index, we present two different
methods for calculating such an index, and several different methods for aggregating these
indices. These indices are proposed as initial efforts that could be implemented very quickly, but
would need to be updated on a regular basis as biofuel production processes, the ability to
measure environmental performance, and environmental goals all advance.
We propose just two measures of environmental performance, GHG emissions and a simple
feedstock rating.
4.1 A quantitative compensatory index
The first example is a simple compensatory index that uses two measures of environmental
performance, GHG emissions and feedstock production practices, and for which better
performance is indicated by higher values. GHG emissions are quantitative and can be
determined by modeling ( for feedstock production) and observation ( for fuel processing).
Evaluation of feedstock production is based on the following categories:
! " conventional row crops, residues, and wastes
! " low- environmental- impact row crops
! " perennial crops
! " low- environmental- impact residues and wastes
Of course, these categories would have to be defined in more detail, and the protocol used to
measure GHG emissions would also have to be defined.
Such a simple index is readily understandable and allows for improvements that innovation
might bring because fuels produced with new technologies could always be assigned higher
ratings. One complication is that “ better performance” often means less of something, like lower
GHG emissions or less soil erosion. To account for this, such measurements should be included
as a negative value, but for understandability the reported value should be positive. Below, we
show one approach to doing so.
The quantitative index could be turned into a simpler rating system, again to improve the
understanding of the differences among different biofuels. One approach would be to use
something like the metals associated with Olympic prizes ( bronze, silver, gold). However, this
approach might limit the number of categories to three and doesn’t easily accommodate ever-improving
performance. A better approach might be to use something like a “ star” rating, where
more stars means better performance. We show examples of both.
A simplified compensatory approach would be quite practical in that much of the information
needed to rate all biofuels is already available. In addition, this approach addresses many of the
environmental impacts that are not already managed through regulation. For instance, water
consumption and emissions by biofuel processing plants, as well as emissions of criteria air
pollutants and toxins, are regulated by local, state, and national laws and permitting systems.
Creating Markets For Green Biofuels 25
GHG emissions, however, are not, nor are many of the environmental impacts of agricultural
production.
In a compensatory index, of course, it is possible to trade off performance on one measure
( GHGs) for performance on another ( feedstock production practices). The overall index can be
readily imagined as a weighted average. However, we need to account for the fact that lower
GHG emissions are better, but that we want to have positive values for the rating. One way to do
this would be to define GWI as global warming intensity, M as the maximum GWI we would
ever expect, a Feedstock Rating value, and weights ! 1 and ! 1 for global warming and feedstock
production, respectively. Then we could calculate the green biofuels index like this:
Green Biofuels Index M GWI Feedstock Rating 1 2 & # ( % ) $ #
In this index, two measures, one for global warming intensity ( GWI) and another for feedstock
environmental impacts, are traded for each other depending on the weights ! 1 and ! 2 that are
assigned.
To illustrate how this index might work, we will use the cases described in Section 6, for which
GWI values are calculated, and to which Feedstock Ratings can be applied. In order to have an
index that has higher values for better performance, we use the following ratings:
Category Rating
Conventional row crops, residues, and wastes 1
Low- environmental- impact row crops 2
Perennial crops 3
Low- environmental- impact residues and wastes 4
If we assume that approximately equal weights for the two components are appropriate, then the
values for M, ! 1 and ! 2 should be chosen so that biofuel production that spans the full range from
worst to best performance under current conditions will affect the green biofuels index about the
same. If we measure the global warming intensity in units of grams of CO2- equivalent per MJ of
fuel, then most biofuels would measure at about 95 or less. For convenience, M could be set to
100, and then appropriate values would be ! 1= 1 and ! 2= 25. Thus, our cases would be rated as in
Table 1.
The table also illustrates a categorical rating system awarding a star for each 40 “ points” in the
combined index. The numerical index has a risk of suggesting that 100 is a perfect score, but the
star rating makes it less likely, as ( for example) generals in the army top out at five stars, but
Michelin restaurant ratings stop at three and hotels at four.
Creating Markets For Green Biofuels 26
Table 1: A Quantitative Compensatory Green Biofuels Index ( illustrative values)
Case Fuel / Technology GWI Feedstock
Rating
Green Biofuels
Index
Value Star
Rating*
1 Conventional agriculture,
coal- fired dry mill with CHP
93 1 ( 100- 93) + 25 = 32
2 Conventional agriculture,
natural gas
65 1 ( 100- 65) + 25 = 60 !
2a Improved corn agriculture,
natural gas
65 2 ( 100- 65) + 50 = 85 ! !
3 Conventional agriculture,
natural gas, no drying
56 1 ( 100- 56) + 25 = 69 !
4 Conventional agriculture,
biomass gasification
42 1 ( 100- 42) + 25 = 83 ! !
4a Improved corn agriculture,
biomass gasification
42 2 ( 100- 42) + 50 = 108 ! ! !
5 Switchgrass 16 3 ( 100- 16) + 75 = 159 ! ! !
6 Low- environmental- impact
residues and wastes
25 4 ( 100- 25) + 100 = 175 ! ! ! !
* One star is awarded for each 40 value units.
Global Warming Intensity
A GWI measure should be associated with each batch of biofuel, based on the combined GWI of
the feedstock and biorefining phases. Initially, this measure would combine an average global
warming intensity per distinct feedstock, as estimated by a transparent, publicly available model
( e. g. GREET) based on average feedstock production methods, with the calculated results for a
specific biorefinery.
The standard would initially address only the three main greenhouse gases, carbon dioxide
( CO2), methane, ( CH4), and nitrous oxide ( N2O), weighted by their 100- year global warming
potentials as per the latest available IPCC assessment, currently 1, 23, and 296, respectively.
While we recognize that other factors ( e. g., aerosols, particulates, albedo) affect the climate
impacts of producing and using biofuels, these factors are less well understood and are included
in only one model, which is not publicly available ( Delucchi 2003). The purpose of measuring
GWI is not to determine a “ true” value; this is not possible given the many uncertainties
involved. Rather, the goal is to create a transparent estimation of the impacts that is accurate
enough to create incentives for lower- GWI production methods. Accounting for the three
primary GHGs achieves this goal.
Note that GWI should be reported as an absolute measure for each biofuel pathway. It is
tempting to report the GWI of biofuels relative to gasoline ( i. e., as a percentage of gasoline’s
GWI), but to do so creates confusion, because there are many possible formulations of gasoline
( conventional, California reformulated, gasoline “ blendstocks”), numerous feedstocks ( e. g.,
petroleum, tar sands, extra heavy oil, coal), and many processes ( e. g., conventional on- shore
production, off- shore production, enhanced recovery, coal liquefaction) with differing GWIs.
Moreover, the average GWI of gasoline is increasing as lower- quality resources are exploited.
Instead, the GWI of biofuels should be understood as an absolute measure that can be compared
to the measure of gasoline, and of diesel, and electricity, and all other transportation fuels.
Creating Markets For Green Biofuels 27
While it is important that a green biofuels standard be used to encourage better environmental
outcomes among biofuels, it is equally important that biofuels not be disadvantaged vis- à- vis
other fuels because of a green standard. Instead, every effort should be made to ensure that
standards equally applicable to various fuels be equally applied, to encourage better
environmental outcomes for fuels. The global warming intensity of transportation fuels is
applicable to all fuels, and it should be so applied. This is the only meaningful use of this
measure.
Feedstock Rating System
At this time, it is impractical to measure and track site- specific agricultural impacts such as soil
erosion or nutrient and pesticide runoff. As discussed earlier, we believe that practice- based
standards are more appropriate for agriculture until agro- environmental modeling is able to
provide accurate, robust quantification of actual performance. However, because practices are
not generally amenable to quantification, qualitative ratings should be used. Initially, average
per- feedstock values can be used, allowing feedstock producers to opt in if robust and practical
feedstock- specific measures become available. More refined estimates for specific regions, soil
types, and agronomic practices could be generated, but this lack of data is not an impediment to
creating a useful biofuels standard.
To manage the environmental effects of feedstock production, we propose using a framework
like that created by USDA Organic, the Conservation Security Program, or the Forest
Stewardship Council ( Forest Stewardship Council 2006). These programs design ecological
management plans appropriate to each farm, and use a conservation systems approach rather than
addressing single practices.
4.2 A qualitative compensatory index
Another compensatory approach is illustrated in Figure 3. Here, we simply plot the cases
according to category of feedstock production practices ( horizontal) and total GWI ( vertical) and
designate five regions of performance: unrated and then “ one- star” through “ four- star.” Note that
in this rough, categorical index, many different types of fuel have the same rating and thus may
be indistinguishable. Thus, cases 1 and 3 are both unrated even though they are actually rather
different, and similarly 4 and 2a are both two- star.
Another approach would be to label cases with both the GWI and the overall rating, so Case 2
would be ! – 65, while Case 4 would be ! ! – 42. However, this gives further weight to the GWI.
It might be simpler just to have a dual rating: a star rating referring to the feedstock production
practice rating and the number referring to the GWI.
Creating Markets For Green Biofuels 28
Figure 3: A Qualitative Compensatory Green Biofuels Index ( illustrative)
4.3 A lexicographic non- compensatory index
It is possible to aggregate the quantitative and qualitative measures to produce a single overall
qualitative rating. However, creating qualitative labels requires defining arbitrary boundaries
between values that naturally occur on a continuum, creating biases toward one pathway or
another. This reduces incentives for producers to continually improve their ratings since no
additional benefits accrue until and unless their product crosses into the next rating category.
A proliferation of numerical ratings, however, would be undesirable. In practice we believe only
one numeric measure must be reported: global warming impact. No biofuel pathways— present
and currently envisioned— use much petroleum ( Farrell et al. 2006; Wang, Wu, et al. 2006;
Wang 2006). Petroleum use should be measured to prevent backsliding, but it need not be
reported to the consumer, nor are incentives or regulations required to improve this outcome.
For reporting overall biofuel environmental performance, we propose a four- tier lexicographic
( non- compensatory) index: Gold, Silver, Bronze, and Brown. The technical definition of these
categories should be adjusted on a regular ( e. g., five- year) basis to incorporate new scientific
understanding, technical capabilities, policy goals, and market conditions.
Though it is beyond the scope of this study to define the specifics of this standard, we offer a
( somewhat incomplete) sketch of how this might work. This is shown in Table 2. Each fuel is
assigned the rating for which it meets all of the applicable standards in Feedstock, Processing,
and GHGs, as shown below. Note that the use of CSP contracts is only one possible approach.
Environmental Performance of Feedstock Production
Environmental Performance of Fuel Processing ( 100- GWI)
Case 1
Case 2
Case 3
Case 4
Case 5
Case 2a
Case 6
Case 4a
Creating Markets For Green Biofuels 29
Table 2: A Lexicographic ( Non- Compensatory) Green Biofuels Index ( illustrative)
Performance Rating Requirements
Feedstock Processing GHGs*
- Post- recycling biogenic waste, or
- Agricultural residues removed from crops
under a Tier III CSP contract
- Zero process water effluent, and
- Maximum of 3.0 gallons of water
consumed per gge fuel ( for
ethanol)
< 40
- Crops under a Tier II or III CSP contract,
or
- Post- recycling biogenic waste, or
- Agricultural residues removed from crops
under a Tier II or III CSP contract
- Zero process water effluent, and
- Maximum of 3.5 gallons of water
consumed per gge fuel ( for
ethanol)
< 60
- Crops under a Tier I, II, or III CSP
contract, or
- Post- recycling biogenic waste, or
- Agricultural residues removed from crops
under a Tier II or III CSP contract
- Low process water effluent, and
- Maximum of 4.5 gallons of water
consumed per gge fuel ( for
ethanol)
< 80
- Crops or residues from farms without CSP
contracts
- Higher- than- benchmark process
water effluent or overall water
consumption
< 100
EXCLUDED
- Feedstocks from converted high- habitat-value
land
- Municipal solid waste before removal of
recyclable material
- Tires, plastics, or fossil- based wastes
> 100
* GHGs are measured in g- CO2 eq/ MJ
Under this system, “ good” environmental performance is variously defined by the “ Bronze,”
“ Silver,” and “ Gold” designations, while the “ Brown” designation is intended to capture fuels for
which no preference over baseline fuels ( e. g., gasoline) is provided. Finally, the system should
establish a minimum level of performance below which fuels would be excluded from any
rating. Such fuels could be prohibited from consumption because of unacceptable environmental
performance.
4.4 Blending fuels and feedstocks
Only the quantitative portion of any index can meaningfully be blended into an average score; a
series of lexicographic ratings cannot be “ averaged” without explicit tradeoffs of non-commensurate
value that the categorical system was originally intended to avoid. Nevertheless,
as would probably be necessary under any regulatory approach that placed requirements on a
blender’s entire product line, arbitrary blending rules can be created to establish that, for
instance, one part gold and one part bronze make a silver. Unrated biofuels, because their
performance has no known minimum value, should not be thus “ redeemed,” however.
BROWN
BRONZE
SILVER
GOLD
Creating Markets For Green Biofuels 30
5 Implementations of a Green Biofuels Index
A wide range of environmental indices for biofuels is possible, varying across several
dimensions, including number of components ( e. g., agricultural practices, GHG emissions, etc.),
accuracy, implementation cost, auditability, and theoretical grounding. Which of the many
possible indices is best depends on the goals the index is intended to support. Government
policies to increase the use of green fuels can be chosen from a surprisingly wide variety of
generic options. Different indices for a particular fuel can vary in many ways. They might
include more or fewer components, and might be determined in practice by methods that are
more or less precise and/ or accurate, more or less expensive to implement, and more or less
auditable or defensible on both theoretical and policy grounds.
We now turn to a set of brief characterizations of some distinctive ways in which a green
biofuels index could be used. Options are discussed in approximately their order of increasing
intrusiveness. For each implementation, we mention how effective it might be in creating
markets for green biofuels.
5.1 Allow the market to find its way to efficient labeling and claims
The least intrusive approach, much like the regime currently in place, allows sellers to determine
whether and how to communicate the environmental performance of their fuel. Completely
private mechanisms of this kind are not unknown; for example, Good Housekeeping magazine
contracts with advertisers to allow the use of a trademarked seal indicating veracity of
advertising claims. It is possible that participants in the energy market would come to a
collective agreement, effective even without legal force, defining a green biofuels index that they
would all use. However, the dimensionality of the measure and the uncertainty over what to
include and how to aggregate individual measurements almost certainly make it difficult for a
robust, effective metric to arise.
Accordingly, such regimes are often fragile, as evidenced by the continuing confusion over food
health claims. This approach also has a fundamental theoretical defect: It does not account for
benefits to society as a whole that private consumers do not wish to pay for. Reducing global
warming by private action is probably the most complete case of common- property resource
market failure: Anyone’s contribution is diluted by being spread across the entire population of
the planet, so anyone who makes a sacrifice for the common good experiences exactly the same
future as someone who doesn’t, whether or not others do likewise. It might be argued that current
interest in government policy in this area is justified largely on this basis alone.
The market has not produced markets for green biofuels so far, and it is implausible that an
unregulated market would do so in the future.
5.2 Define allowable claims and protocols to support them
A step beyond an unregulated market of claims and measures is government prohibition of all
but a single index in marketing claims. The simplest and most familiar example of this approach
is the system of legal weights and measures: The foot, yard, pound, quart, meter, and other
important units are defined in terms of standards maintained by the National Institute of
Standards and Technology ( NIST), and no others are permitted to be used in trade. Another
Creating Markets For Green Biofuels 31
example is USDA’s establishment of an operational definition of organic, and regulatory
restriction of the word to denote only what is covered by the definition. This public action, a
response to criticism of the growing number of standards and certification programs,
strengthened producers’ certainty and simplified consumers’ purchasing analysis, but it also led
to ongoing controversy as to whether USDA had chosen the right definition.
Any version of indexing discussed above is suitable for a regime of this kind, requiring only a
government agency, such as NIST, or a nonprofit organization or research institution to establish
it and legislation to enact it. A federal requirement would avoid a patchwork system of
inconsistent requirements from state to state. Enforcement of unauthorized labeling or
mislabeling might be through state weights and measures agencies, or by private fraud actions
against market participants.
If index credits are detached from physical product, however, it is not clear that labeling for
consumer choice will be acceptable in view of the long- standing expectation that a label describe
the specific item to which it is applied. This expectation is rooted in our traditional association
between labels and the private consequences of buying and using a good. For instance, the gas
mileage on an EPA sticker may motivate a car buyer to choose a small, efficient car for the good
of the planet, but he reasonably expects it to describe his personal experience, not an efficiency
realized by an unknown mix of other cars while he personally pays for more gas than predicted.
Defining allowable claims about the environmental performance seems like a necessary
condition for healthy markets for green biofuels, but it hardly seems sufficient. Green products
typically capture only a very small share of any market, unless they have no additional cost at all.
5.3 Require environmental labeling
A step beyond defining protocols would have the government require labeling much as the FDA
requires processed food to bear ingredient and nutrition labels. These labels, incidentally, offer a
model of what a multidimensional green biofuels index might look like; the tradeoff between
simplicity of use and precision of match to consumer concerns is obvious.
As with Section 5.2, any of the index forms discussed above could apply. However, mandatory
labeling entails the further establishment of size, location, typography, and more. Presumably
fuels would be labeled at the pump, but it could be necessary ( given that people have little
experience with real differences among motor fuels and frequently misunderstand and misuse the
one measure— octane— commonly displayed) to require advertising to carry labels as well, a
daunting expansion of government intrusiveness and oversight.
These two regimes, focused on consumer decision, impose some important constraints on the
form of the index presentation. The most important is simplicity and transparency; buying the
right motor fuel will never justify even the kind of attention people pay to their food, and an
index with a simple scale and/ or few categories will be essential. Also important is consideration
of the implicit as well as the explicit technical information provided. Most rating systems
incorporate social conventions such as an A or a score of 100 indicating a top or best grade, or
the gold/ silver/ bronze ordering of athletic medals. Establishing a scale that doesn’t cap future
Creating Markets For Green Biofuels 32
performance by running out of headroom ( such as awarding an A rating to today’s highest-performing
fuels) and that meets social conventions for legibility is not a simple matter, though
our example of stars that can be awarded without limit may be one.
Requiring environmental labels is a further step toward the development of markets for green
biofuels, but it also seems insufficient. As noted, higher- priced green products typically capture
only a very small share of any market. And millions of Americans consume unhealthy and even
unsafe products despite warnings and public service advertising.
5.4 Require government ( and contractors) to purchase green biofuels
Government often uses its purchasing power to demonstrate leadership and help develop markets
for products with socially desirable properties. For example, many governments require that all
paper purchased have a minimum recycled content. Similarly, governments could require that
their agencies ( and possibly their contractors) purchase only biofuels with a minimum green
index rating. As purchase is a binary action ( buy or don’t buy), any index used for this option
must be one- dimensional in the end.
Government procurement has a mixed record in supporting the development of new markets for
environmentally preferable products. Successful government markets can support specialized
producers or specialized divisions within larger firms, and these producers’ operations may yield
innovation that can spread, lowering costs and improving performance throughout the industry.
Less successful interventions create high costs for taxpayers and entrenched niche producers
with little public benefits. The size and the direction of this effect are critically influenced by the
size of government procurement relative to the total market, the size of government procurement
within each firm’s sales, and the market structure of the industry ( multiple highly competitive
firms are likely to show larger effects than fewer oligopolistic firms). Green procurement by
government shows the highest benefits when it is directed at innovation rather than the support
of ongoing operations. Based on these criteria, government procurement standards for green
performance in the nascent biofuels industry could exert some influence on innovators, but it is
unlikely to shift the industry significantly.
5.5 Subsidize or tax based on environmental performance
Expanding the scope of market intervention beyond government purchases, government could
pay direct subsidies at varying levels according to an environmental index, or tax fuels according
( most simply) to their net GHG emission. This policy is analogous to the current ethanol subsidy
but could be much better targeted and more efficient in diverting the market to better fuels. In
theory it is possible to determine optimal tax rates by measuring the costs and benefits of fuel
production accurately and unambiguously. However, for the environmental performance of
biofuels, this is likely to be impractical because of problems associated with measuring the
physical changes from agricultural production. Moreover, the subjective judgments required to
monetize these changes ensure that taxes or subsidies will be far from perfect. Consider the
fierce debate about just one issue, climate change damages, as illustrated with the recent Stern
report ( Stern, Peters, et al. 2006). The decision to tax environmental externalities is not
Creating Markets For Green Biofuels 33
avoidable: A tax of zero is a tax like any other, and obviously it misrepresents the social cost of
individual behavior.
Payments ( either taxes or subsidies) provide for flexibility. Producers can choose the level of
environmental performance that is efficient for their businesses. Furthermore, taxes create a
pervasive incentive for all producers to find ways to do less of the taxed behavior at all times,
whereas a regulation provides no such incentive once compliance is achieved, and low- cost
emission reducers become an important political constituency for the policy. Also, payments can
be adjusted to follow changes in biofuel production and environmental goals. Finally, a tax
mechanism is much more flexible and adaptable to a multidimensional measure of environmental
benefit; each dimension can be assigned its own tax rate. Regulation, on the other hand, typically
demands high aggregation or else risks completely ignoring important dimensions. Accordingly,
a tax or subsidy scheme allows the most complete and accurate incorporation of the index
information available of any of these options.
This option, along with the two that follow, impose the least burden of analysis on consumers,
and best protect individuals from the perverse incentives of the common property resource
problem.
Because market participants are focused on costs and prices, the use of subsidies or taxes could
strongly support the development of markets for green biofuels.
5.6 Require an aggregate green biofuels performance
Mandating the environmental performance of an overall industry is likely to ensure a specific
environmental outcome while preserving some flexibility for producers to meet the overall
standard. The Corporate Average Fuel Economy ( CAFE) requirements for automobiles is a
policy in which sellers of a product are obliged to maintain some average performance level in
their total sales.
As discussed in Section 5.5, above, even the most complex indexing information can be
incorporated into a requirement of this kind. The main disadvantage of this approach, and the
one that follows, is its implicit acceptance of an infinite step in the marginal benefit schedule: A
prohibition, in practice, means that something below it is so bad on the dimensions constrained
that it can’t matter what other benefits might flow from a small shortfall. A related disadvantage
is the inability of a prohibitory regime to display or encourage improved performance above the
minimum demanded.
Fuel producers are already regulated in many ways, and these regulations have changed fuel
markets substantially. Regulatory requirements for environmental performance are likely to have
a similarly strong effect. One example of such an approach is the Low Carbon Fuel Standard
being developed in California ( Schwarzenegger 2007), which would encourage a market for
fuels with lower GWI. This is both broader than the markets for green biofuels envisioned here
( because other fuels, like electricity, could compete) and narrower as well ( only GWI is
considered.
Creating Markets For Green Biofuels 34
5.7 Forbid sale of fuel below some level
The most coercive policy alternative is to simply forbid the production of fuel whose
environmental index is below a prescribed level. For the near future, the available quantities of
biofuels and other non- fossil fuels, even with extremely optimistic assumptions, suggest that a
policy of this kind is impractical before a long period of adaptation and capital investment. Such
an approach would create markets for fuels with some minimum level of environmental
performance, but not necessarily for greener biofuels. Where the risks of certain practices are
extremely high, for instance in the loss of both tropical rainforest and peat soil carbon in the
conversion of palm oil plantations in Indonesia, outright bans may be appropriate.
Creating Markets For Green Biofuels 35
6 Case Studies
The following case studies demonstrate the practice measures that would be applied to feedstock
production and the quantitative measures applicable to biorefining. We then illustrate how these
feedstock and biorefinery ratings could be combined into a single rating. We present six cases
that demonstrate how practice measures could be applied to feedstock production and the
quantitative measures applicable to biorefining. Five are for various types of corn ethanol, and
the last is for ethanol based on switchgrass or corn stover.
6.1 Feedstock production
As described above, a biofuels index will need to use practice- based measures to identify and
encourage ecologically preferred feedstock production systems. To demonstrate the range of
outcomes, we examine three ethanol feedstock production systems: best- practices corn
production, switchgrass, and conventional corn production with stover collection.
6.1.1 Best- practices corn
The Willow Creek Farm produces corn and soybeans on 3,800 acres in southwest Minnesota.
The farm is operated under a Conservation Stewardship Plan, qualifying it for a CSP Tier III
contract, developed with a USDA technical service provider. The plan identifies the unique
resources and constraints of the farm and identifies the specific practices to be followed to
minimize impacts. The crops are grown in rotation to reduce fertilizer and pesticide needs. A
“ ridge till” tillage system and filter strips on downslope field edges reduce erosion and runoff.
Soil and crop tests are used to determine fertilizer needs before and during the season, and
tractors are outfitted with a Global Positioning System ( GPS) allowing precise placement of
seed, fertilizer, and pesticide.
Corn and soybeans produced by Willow Creek receive a Silver rating, or three stars, the top
ratings available to annual row crops. To understand why this corn does not earn a Gold rating,
it’s helpful to compare corn to switchgrass. The corn receives about 170 pounds of nitrogen per
acre, while switchgrass is expected to require between 50 and 150 pounds per acre. Corn is also
an annual crop, requiring replanting every spring. Switchgrass is a perennial that is replanted
approximately once a decade. So