The workshop’s first session offered an overview of the major issues related to biofuels in order to provide a common starting point for the presentations that would take place during the remainder of the workshop’s 2 days. The issues reviewed in this session included the impacts of biofuels use on greenhouse gas emissions, water quality, and land use; the connection between increasing biofuels production and food insecurity; and the economics of biofuels production.
The session’s first presenter was Timothy D. Searchinger, a research scholar and a lecturer in public and international affairs at Princeton University’s Woodrow Wilson School. He spoke about the implications of biofuels use for the level of greenhouse gases in the atmosphere and, thus, the potential implications of biofuels use for global warming. A secondary topic was the effect of biofuels use on the worldwide food supply and how an increased emphasis on biofuels could bear on world hunger.
Liquid biofuels used for transportation represent just one form of bioenergy, Searchinger noted, but at this point it is the major form. About 40 percent of the U.S. corn crop is used to make ethanol, for example, and there is almost a comparable amount of ethanol being made
1 The planning committee’s role was limited to planning the workshop, and the workshop summary has been prepared by the workshop rapporteur as a factual summary of what occurred at the workshop. Statements, recommendations, and opinions expressed are those of individual presenters and participants, and are not necessarily endorsed or verified by the Institute of Medicine, and they should not be construed as reflecting any group consensus.
from sugarcane in Brazil. Europe produces somewhat less biofuels, although still a considerable amount, mainly in the form of vegetable oil for biofuel.
Many governments around the world have either goals or mandates for biofuels, Searchinger said, and if these goals and mandates are met, biofuels will account for about 10 percent of the world’s transportation fuels by 2020. This represents about 2.5 percent of the world’s total energy budget, but Searchinger said, when the energy that it takes to make biofuels is taken into account, biofuels would be providing about 1.7 percent of the world’s delivered energy by 2020.
How much of the world’s crops would that take? By 2020 biofuels would require that about 26 percent of all the energy contained in the present production of the world’s crops. By 2050 that figure would rise to 36 percent, he added. “So, that gives you some idea of the challenge, which is that it takes a large amount of biomass to get a small amount of energy.”
Of course, liquid biofuels are only one form of bioenergy that people are interested in, he noted. For example, there is also a big push in Europe as well as in some U.S. states to produce electricity from wood products.
Governments are encouraging the use of bioenergy in various ways. The European Commission has required, for instance, that 20 percent of all energy in Europe be renewable by 2020—not just the energy from utilities, but all energy. It is expected now that more than half of that will come from bioenergy, Searchinger said. A number of states have renewable energy targets, he said, although they are not quite as stringent and are just for electricity.
The Effect of Biofuels Usage on Carbon Dioxide Levels
One of the main reasons that people support the use of biofuels, Searchinger said, is the belief that “when you switch from burning a fossil fuel to burning a biofuel you get some kind of direct greenhouse gas benefit.” But, he said, a close examination indicates that this is not the case and that the belief that there is a direct benefit stems from an “accounting error.”
The belief that burning biofuels contributes less carbon dioxide to the atmosphere than burning fossil fuels stems from the fact that biofuels are derived from plants, which absorb carbon dioxide as they grow. “So, the theory is that, in effect, bioenergy is just recycling carbon, not
emitting new carbon.” That is wrong, however, for the simple reason that land typically supports plant growth, whether it is used for bioenergy or not. For bioenergy to reduce greenhouse gas emissions through plant growth, it must lead to additional plant growth.
Searchinger showed a comparison of the net greenhouse gas emissions for gasoline versus ethanol (see Figure 1-1). The first few steps of the production process for biofuels always produce more emissions than the production process for gasoline, he said, “because it takes more energy and emissions to produce the crop and to transform it into ethanol than it does to mine crude oil and refine it into gasoline.” The greenhouse gases emitted from the tailpipe of a car are about the same for ethanol and gasoline, but there is a another source of greenhouse gas emissions for the ethanol, which is the fermentation process. For every 2 grams of carbon emitted from a tailpipe for ethanol, there is another gram produced in the fermentation process.
FIGURE 1-1 Net greenhouse gas emissions for ethanol versus gasoline.
NOTE: CO2 = carbon dioxide.
SOURCE: Searchinger, 2013.
When all of the emissions have been taken into account, ethanol releases almost twice as much greenhouse gas into the atmosphere per unit of energy as gasoline. However, when the corn is grown to produce the ethanol, it absorbs carbon dioxide from the environment, so, in effect, the tailpipe emissions and the fermentation emissions can be ignored because that is carbon dioxide that had been pulled from the atmosphere by the growing plants. With this “plant credit” taken into account, the calculations show that the net greenhouse gas emission for the ethanol is 29 percent less than for the gasoline.
The key concept here, Searchinger said, is that the benefit from ethanol depends on the existence of an offset that makes up for the fact that producing and burning ethanol actually creates much more greenhouse gases than producing and burning gasoline. So, the question is: Is there really such an offset? It is true that growing corn leads to a certain amount of carbon dioxide being pulled from the atmosphere, but that is not all that goes into the determination of the offset. The critical requirement for an “offset” is that it be additional. No one can take credit for a carbon sink, such as a tree if that tree already exists anyway—in this case, regardless of whether the biofuels exist or not. One must take into account all of the circumstances surrounding the production of the ethanol and compare what happens when corn is being grown to produce ethanol to what happens when corn is not being grown to produce ethanol.
The first thing that must be considered is the land that is used to grow plants. “Land grows plants whether it’s growing those plants for biofuels or not,” Searchinger pointed out. “So, those plants are already up taking carbon if you’re growing it for biofuels or not.” Thus, the only way that there is a legitimate offset from growing corn for ethanol is if more plants are being grown on that same amount of land or, specifically, if more carbon dioxide is taken up by the corn crop than was taken up by whatever was growing on that land before the corn. “One way to think about it is that if you had a bare piece of land and you allowed it to grow as a forest, that forest would accumulate carbon, and it would reduce greenhouse gas emissions. On the other hand, if you simply had a forest that was growing anyway, you couldn’t count that as an offset.”
Ignoring this basic fact is a fundamental error that often appears in calculating the biofuels offset. “Biofuel analysis assumes typically that all plant growth offsets biofuels, rather than only additional plant growth,” Searchinger said.
There are various ways that a real offset can be achieved, Searchinger noted. If, for example, a corn crop is planted on land that had few or no plants growing on it before, that is a legitimate greenhouse gas saving because there is a new crop absorbing additional carbon dioxide. Similarly, if there is a crop residue, such as rice straw, that would normally be burned but instead is used to make ethanol, that is a legitimate greenhouse gas saving because the usual carbon dioxide emissions from that rice field have been reduced. On the other hand, if one takes a corn crop that would have been grown anyway and simply diverts it for use in fuel, that does not represent a legitimate offset because there is no additional plant growth and no additional carbon dioxide removed from the atmosphere.
If biofuels use crops from existing cropland, there are indirect ways it could generate the offset and therefore greenhouse gas saving. If, for example, the corn crop used to produce ethanol is diverted from human and animal consumption—and assuming that the corn is not completely replaced by something else—then people and livestock eat less, and they consequently release less carbon in the form of greenhouse gases, mainly carbon dioxide and methane. “Most of us would not think that’s a good thing,” he noted drily, “but that does give you a greenhouse gas benefit.”
Another type of indirect effect on greenhouse gases would arise if an increased demand for corn led farmers to increase their yield, growing more corn per acre, and thus absorbing more carbon dioxide per acre than before. That offset might be reduced somewhat, Searchinger noted, by higher greenhouse gas emissions from increased fertilizer use.
A third type of indirect effect of adopting ethanol would be the plowing up of new land in order to grow corn where no crops had been grown before. The corn crop will absorb carbon dioxide, but whatever plants had existed on the land before are no longer absorbing or continuing to store carbon dioxide, and one must compare the carbon absorbed by the corn with the carbon released (or not subsequently sequestered) by plowing up forests and grasslands.
Searchinger offered an analysis of the costs and benefits of creating new croplands from forests or grasslands. A hectare of corn that is grown for ethanol has a net benefit of about 3 tons of carbon dioxide per year. That counts the benefit of reduced emissions from gasoline, but includes the greenhouse gas costs of producing the corn and refining it into ethanol. But it also has another cost, which is the opportunity cost of the land, or the carbon benefits the land would produce if not used for ethanol. If that same hectare of land in Iowa were to be left fallow and
allowed to reforest, it would absorb 7.5 to 12 tons of carbon dioxide per year, so using the land to grow corn has a net cost in terms of the amount of carbon dioxide being sequestered. The most effective approach would be producing cellulosic ethanol—ethanol produced from wood or grasses—on the land, but even in that case, Searchinger said, “you’re simply matching the opportunity cost of using that land for another purpose.” And if the cropland is created by clearing forest, there is a much greater cost in greenhouse gases because plowing up forests will release 12 to 35 tons of carbon dioxide per hectare each year for 30 years. Thus, the best-case biofuels scenario would be to take fallow land and use it for the production of cellulosic ethanol, he said, but even in that case it is only a break-even situation if the land would otherwise come from abandoned land, and it would increase emissions if the land used was previously forest. There is no offset.
In short, out of the three possible indirect effects of growing corn to produce ethanol, Searchinger said, two are bad. “Land-use change leads to greenhouse gas emissions and habitat loss, and reduced food consumption leads to hunger. The only real effect that would be beneficial is if the overwhelming response for biofuels was simply that farmers produce more food on the same land.” The implication is that in deciding to create biofuels simply by going out and buying crops that were already growing, policy makers are, in effect, betting that the main effect of this policy will be a yield gain and that there will be relatively little decrease in food consumption and relatively little increase in the amount of land devoted to farming.
It is not at all clear that this is what has happened, Searchinger said, and he then took a closer look at the effects of biofuels production on food consumption.
Biofuels and Food Consumption
Interestingly, although reduced food consumption would not appear to be a desirable result, it is exactly what is assumed in the major models used to predict the greenhouse gas effects of biofuels, Searchinger said. “You have to find this deeply in the data,” he said. “It’s generally not reported. Take, for example, the Environmental Protection Agency [EPA] analysis of corn ethanol, which found relatively little land-use change compared to some other studies. One reason it didn’t find as much land-use change as other studies is that it actually estimated that a quarter of all the calories that are diverted to ethanol aren’t replaced.”
Similarly, the model used by the California Air Resources Board assumed that more than half of the calories from the corn diverted from human and animal consumption to ethanol would not be replaced. A major model used by the European Union assumes that a quarter of the calories from either corn ethanol or wheat ethanol are not replaced.
Thus, the greenhouse gases benefit from using biofuels, as calculated by these models, depends on humans and animals eating less, expending less energy, and thus breathing out less carbon dioxide (and producing less methane). “If you were to eliminate these savings,” he said, “you would not have greenhouse gas savings according to all these models.”
Of course, he noted, the decreased consumption assumed by these models is not a desirable effect because there remains a great deal of hunger in the world—roughly 900 million people are hungry according to recent estimates. Thus, it is particularly worrisome that the frequency of food crises worldwide has essentially tripled since 2005, when the amount of biofuels use began to increase sharply. And according to a recent report by the High-Level Panel on Food Security, of which Searchinger is a member, that is not a coincidence (HLPE, 2013). “We basically conclude that biofuels are the dominant source of food price increases.”
In particular, the increase in corn prices in the United States can be traced to the cost of oil combined with government tax credits for ethanol production. With crude oil at $80 per barrel and with the current U.S. tax credits for ethanol, it is economical to use corn to make ethanol and to replace gasoline until the price of corn reaches about $6.80 per bushel. “Roughly speaking,” he said, “this is a 275 percent higher price than the long-term corn price in the first part of the 2000s.” Thus, corn prices get bid up until they get close to that level—and as the price for corn intended for ethanol production increases, the price for corn intended for consumption increases along with it, for the crops are the same. Furthermore, as the price of corn increases, the price of wheat and soybeans—and, to a lesser extent, rice—track the price of corn very closely because the crops can, to a significant extent, be substituted for one another. “So, this force by itself is perfectly adequate to explain the vast majority of the price rise that we’ve had,” Searchinger said.
Some people have suggested that the increase in prices have been due mainly to supply problems, but Searchinger said that grain yields have increased steadily, so “the problem was not that supply wasn’t doing its best to keep up.” Instead, the problem has been the rapidly growing demand for grains for use in biofuels. “Although farmers have
been able to keep up with the growth of demand for food and feed, farmers haven’t quite been able to keep up with that rate of growth and to supply biofuels at the same time.”
Global Consequences of Increasing Biofuels Use
In the last portion of his talk, Searchinger took a broad look at the environmental costs of biofuels making up a large percentage of transportation fuels or, more broadly, bioenergy making up a large percentage of the global energy budget.
He began by asking whether the yield of key crops could be expected to increase enough over the next several decades to produce the necessary biofuels without adding additional cropland. An examination of the trends in crop yield from 1960 to 2006 shows that there was a steady, linear increase during that time that changed little from year to year. During that time, the crops were devoted mainly to food and feed, but the growing demand for biofuels in the future will require that yields be increased more quickly than in the past. To produce all the crops needed by 2020 for both food and biofuels without any change in land use will require a doubling of the historical yield growth rate, Searchinger said, “and that’s not going to happen.”
Compounding the problem is that in addition to the growing demand for biofuels, there is projected to be a huge growth in demand for food between now and 2050. “You need to produce 64 percent more crop calories by 2050 compared to 2006, and 70 percent more meat and milk,” he said. “When you compare that to historical growth rates, that means calorie growth would have to grow as much as it did each year during the height of the green revolution, and meat production would have to grow 30 percent higher.” Furthermore, even during the green revolution, when irrigation rates doubled and many parts of the world began using fertilizer that hadn’t used it before, there was still an increase of more than 400 million hectares of land devoted to crops. “So, the point is that even without an increased demand for crops for biofuels, it will be a very tight challenge to produce all the food we want on the same land,” Searchinger said.
Thus, there will certainly need to be land-use change if the demand for biofuels is to be met in the future, and a number of studies have examined the issue of how much bioenergy could be
created by devoting land to crops for bioenergy. One study by the Intergovernmental Panel on Climate Change, for example, asked how much bioenergy could be produced by using all the world’s potential cropland that is not being used for crops. The study found that there were about 1.3 billion hectares of potential cropland that could be used for bioenergy. “If we use all of that for biofuels and get really high yield,” Searchinger said, “we could essentially produce all of our energy needs with bioenergy.” But the analysis ignored the carbon costs of the change in land use.
Some of those 1.3 billion hectares are forests, which store and in many cases continue to accumulate carbon dioxide. Indeed, Searchinger noted, “the accumulation of carbon in forests is a huge factor holding down climate change. It’s why when we burn carbon dioxide we assume that only half of it stays there [in the atmosphere], in part because a quarter of it goes into forests.” So, cutting down forests to produce croplands to produce bioenergy would, in essence, add a great deal of carbon dioxide to the atmosphere.
Another type of land that could be used to produce bioenergy is abandoned cropland. But in many cases that abandoned cropland is slowly returning to forest, so, again, turning it to cropland would have the effect of adding carbon dioxide to the atmosphere.
Yet another type of land that is talked about as potential cropland are grazing lands and savannas, particularly in Africa. Searchinger and colleagues have prepared a paper analyzing whether there would be a greenhouse gas benefit by using the African savannas to make biofuels. Assuming “a pretty high yield,” the group estimated that “only about 3 percent of that [savanna] would produce a greenhouse gas benefit over 20 years.”
The real challenge with bioenergy, Searchinger said, is that photosynthesis is extremely inefficient. “If you’re really lucky you get half a percent of the solar energy transformed into plant biomass—that’s extraordinary achievement over the course of the year. And eventually maybe a tenth or two-tenths of the original solar energy will end up actually in delivered energy like electricity.” By contrast, a solar cell turns 10 percent of solar energy into electricity. “So, compare one-tenth of 1 percent with 10 percent, and you’ll get an idea of the inefficiency of using land. What that means is it takes a tremendous amount of land to make a small amount of bioenergy.”
The bottom line is that to provide 10 percent of the world’s transportation fuel by 2050 would require 36 percent of all of today’s crop production, and it would amount to less than 2 percent of the world’s delivered energy at that time.
There are some people who have a much more ambitious goal for bioenergy—they would like to produce 20 percent of the world’s energy from bioenergy by 2050. That would require all of the plants harvested today around the world for any purpose—all crops, all grasses eaten by livestock, all wood, and all crop residues.
The bottom line, Searchinger said, is that trying to make energy from crops will drive up food prices and have a major health effect through hunger, while at the same time, because of the issues surrounding changing land use, increasing biofuels production would not actually reduce greenhouse gas emissions.
In the session’s second presentation, Howard Gruenspecht, the deputy administrator of the U.S. Energy Information Administration (EIA), placed biofuels in the context of the larger energy picture. The EIA, he noted, is a statistical and analytical agency within the Department of Energy, and, as such, it aims to provide an objective view of biofuels and not to advocate for one position or another. He began by discussing the current role of biofuels in the overall energy picture and then spoke about the outlook for biofuels in the future.
The Current Role of Biofuels
Although the use of liquid biofuels has grown significantly during the past decade, Gruenspecht said, biofuels still represent a very small part of the nation’s overall energy use. In particular, as can be seen in Figure 1-2, they currently provide about 1 percent of total U.S. energy. They represent a somewhat larger but still modest share—about 4 percent by energy content—of the supply of transportation fuels.
The role of biofuels can be considered from various perspectives in addition to their percentage of the total U.S. energy budget, Gruenspecht said. For example, one of the motivations in adopting policies to increase the use of biofuels has been concern over how much the country relies on imported oil. From that perspective, a key fact is that the country’s reliance on oil imports has been declining sharply during the past 8 years. In 2005 net petroleum and biofuels imports peaked at about 60 percent of the total liquid fuels used in the United States; by 2012 that had
FIGURE 1-2 Sources of U.S. energy, 1980–2012.
SOURCE: EIA, 2013.
dropped to only about 40 percent, he said, and the EIA’s latest short-term outlook indicates that this will have fallen even further, to slightly above 30 percent, by 2014. “Higher production and lower consumption are both contributing to recent trends in this area,” he said.
The transportation sector alone accounts for 70 percent of total U.S. liquid fuel use. Gasoline now accounts for about 60 percent of all liquid fuels used by the U.S. transportation sector, Gruenspecht said, and that gasoline contains an average of about 10 percent ethanol by volume. In absolute terms, the use of motor gasoline was growing steadily until about 2005, at which point it leveled off for several years and then began falling. This decrease in gasoline consumption is part of an overall decline in energy consumption by the transportation sector, which is in part due to the effects of the recent economic downturn, which has led to a decrease in vehicle miles of travel. Another factor in the decline in energy consumption by the transportation sector has been the growing efficiency of the light-duty vehicle fleet, which has resulted both from market effects and regulations on fuel economy. “More stringent fuel economy regulations that have already been promulgated for light-duty vehicles through model year 2025 are expected to have a further impact in dampening the demand for gasoline,” Gruenspecht added. This is vital
context for biofuels, he noted, because at present about 98 percent of all ethanol being produced in used as a blend stock in gasoline.
Trends in Biofuels Use
Taking a closer look at the growth in biofuels, Gruenspecht showed how biofuels use has grown during the past decade. In 2000 ethanol made up well under 1 percent of the gasoline fuel sold in the United States. That number began climbing steadily around 2002 and started to accelerate around 2005 or 2006. By 2011 ethanol made up more than 9 percent of all gasoline sold by volume, although it represented only about 6 percent of gasoline fuel in terms of energy content. By contrast, biodiesel makes up a much smaller percentage of all diesel fuel sold in the United States—just fewer than 2 percent both in terms of volume and in terms of energy content.
Ethanol is not likely to move past 10 percent of total gasoline sales (by volume) any time soon, Gruenspecht said. “Although the EPA has recently approved the use of gasoline blends containing up to 15 percent ethanol in all model-year 2001 and newer light-duty vehicles,” he said, “10 percent remains the ethanol blending limit for gasoline used in older vehicles as well as in marine applications and for small gasoline engines used in power equipment like your lawn mower or chainsaw.” Furthermore, the EPA’s approval of 15 percent ethanol blends for use in newer vehicles has been controversial, he said, and there has been little movement toward widespread retail distribution of blends with more than 10 percent ethanol. Thus, 10 percent is often referred to as the “blend wall,” meaning that it is represents a level that will not be easy to break through.
In economic terms, ethanol has generally been somewhat less expensive than gasoline on a per-gallon basis during the past several years. However, global crude oil, gasoline, and diesel prices were all at record high annual average levels in 2012, and even given this favorable situation, ethanol producers were seeing relatively little profit, thanks in large part to the high prices for corn.
Furthermore, if prices are compared on the basis of energy content, ethanol has been consistently more expensive than gasoline because a gallon of ethanol has less energy than a gallon of petroleum-based gasoline. Thus, from this perspective, ethanol producers are making little
money even though their biofuel is more expensive than gasoline in terms of energy supplied per gallon.
“The difference has some major implications for the economic prospects of expanding the use of ethanol in higher percentage blends where its lesser energy content per gallon would be more noticeable to consumers,” Gruenspecht said. Years ago, when customers had a choice between 100 percent petroleum-based gasoline and gasoline containing 10 percent ethanol, their behavior indicated that they generally looked for the cheapest price per gallon without taking into account the lower energy content of gasoline with ethanol. And today it does not matter because nearly all gasoline available at retail outlets contains 10 percent ethanol by volume. But things could change with a move to make gasoline with higher ethanol content, he said. “Experience in Brazil, where high-percentage ethanol fuels are sold, suggests consumers definitely make purchase decisions based on energy content pricing rather than simply buying the cheapest gallons when considering such fuels. So, there are some interesting economics that come into play in this area.”
In the case of diesel, biodiesel is significantly more expensive than 100 percent petroleum-based diesel fuel. The main reason it is used today is the existence of tax credits and some mandates requiring a specified percentage of biodiesel in the overall fuel pool, Gruenspecht said.
For much of the past decade, the United States was a net importer of biofuels. This was particularly true from 2006 to 2008, Gruenspecht said, when “the phaseout of other octane enhancing additives used in petroleum gasoline created a sharp upsurge in ethanol consumption.” During that time, much of the imported biofuels came from Brazil.
However, since 2010 the United States has been a net exporter of biofuels, and, interestingly enough, there has been a significant amount of two-way trade with Brazil. “The two-way trade generally involves U.S. imports of sugarcane ethanol and U.S. exports of corn-based ethanol,” he said. In terms of the characteristics of the fuel, the two types of ethanol are indistinguishable, he said, so the United States is both sending ethanol to Brazil and buying ethanol from Brazil. The existence of this unusual two-way trade is “in large part attributable to specific features of federal and state level policies which provide several types of extra credit for using ethanol produced from sugarcane due to calculations related to life-cycle emissions.”
Federal and state policies regarding biofuels are constantly evolving, Gruenspecht said. For example, there was a federal requirement in the early 1990s for the use of oxygenates in reformulated gasoline. It was intended to promote the use of ethanol, but in most markets the non-ethanol oxygenate MTBE (methyl tertiary-butyl ether) outcompeted ethanol. However, after 2000 a number of states became concerned about possible ground-water contamination by MTBE and banned it, which caused ethanol use to grow significantly. Then the Energy Policy Act of 2005 led refiners to decide that the liability risks of continuing to use MTBE were too large, and its use was phased out on a national basis beginning in 2006, which led to an upsurge in demand for ethanol.
Furthermore, since the 1980s there have been federal—and, in some cases, state—tax incentives for the use of ethanol and other biofuels. These tax incentives combined with the phaseout of MTBE drove much of the increase in biofuels use.
Three federal tax incentives—two for ethanol and one for biodiesel—expired at the end of 2011. One incentive, the tax credit for cellulosic ethanol, was scheduled to expire at the end of 2012, but it was renewed as part of the “fiscal cliff” legislation.
The renewable fuels standards, which were enacted in the Energy Policy Act of 2005 and expanded by the Energy Independence and Security Act of 2007, set a series of annual targets for various groups of biofuels. “The number that people remember,” Gruenspecht said, “is the 2022 number for total biofuels, which is 36 billion gallons, of which 21 billion falls in the category of advanced biofuels, with 16 billion of that 21 billion falling in the advanced fuels subcategory of cellulosic biofuels.”
To date, he added, there has been little success in producing cellulosic biofuels, and the EPA has ended up modifying the statutory goal in each of the years through 2012. “So, for instance, in 2012 instead of 500 million gallons, they set a goal of about 10 million gallons.” The 2013 goal is likely to be modified as well, he said.2 “It’s supposed to be a billion gallons. … I don’t know exactly what it’s going to be. I know that my agency is required by statute to send a letter to the EPA each October to say what we think can be produced, and we thought in
2 Since the time of the workshop, the EPA has modified the statutory goal for cellulosic biofuel to 6 million gallons. Available at http://www.epa.gov/otaq/fuels/renewablefuels/documents/420f13042.pdf (accessed August 20, 2013).
October 2011 that up to 10 million gallons could be produced in 2012.” (That estimate has since been revised downward to only 4 million gallons.) The yearly statutory goals are set to rise rapidly through 2022, he noted. By 2016, for instance, the country is supposed to be producing 4.25 billion gallons of cellulosic biofuels. It will be a “tough slog,” Gruenspecht predicted.
The Outlook for Biofuels
With that background, Gruenspecht described the outlook for biofuels over the next decade. Before he began, however, he noted that there is a great deal of uncertainty in the forecast. “Energy projections can be wrong for any number of reasons, including assumptions about economic growth, energy market developments, energy technology improvements, changes in consumer preferences, and energy policy developments, only to name a few.” So, the outlook should be heard with that in mind.
The outlook is based on the EIA’s Annual Energy Outlook 2013. It generally assumes current laws and regulations, he said, and although it does assume technological improvements, they are only trend improvements. “We’re not guessing when breakthroughs will occur,” Gruenspecht said. The key results from the projection are
• Growth in energy production outstrips consumption growth.
• Crude oil production, particularly from tight oil3 fields, rises sharply during the next decade.
• Natural gas production grows faster than in previous projections, serving the industrial and power sectors and an expanding export market.
• Motor gasoline consumption reflects the introduction of more stringent fuel economy standards, while diesel fuel consumption is moderated by increased natural gas use in heavy-duty vehicles.
• The United States becomes a larger exporter of natural gas and coal than was projected in earlier projections.
• All renewable fuels grow, but biomass and biofuels growth is slower than in previous projections.
3 “Tight oil” refers to the light crude oil trapped in shale, limestone, and sandstone formations characterized by very low porosity and permeability.
• U.S. energy-related carbon dioxide emissions remain more than 5 percent below their 2005 level through 2040, reflecting increased efficiency and the shift to a less carbon-intensive fuel mix (EIA, 2013).
Annual Energy Outlook 2013 projects where the United States will get its energy through 2040. As can be seen in Figure 1-3, the expectation is for overall energy use to rise very slowly. Thanks in large part to increasing energy efficiency, primary energy consumption is not projected to get back to the 2007 level until about 2022.
Renewables are predicted to meet a large share of the growth in energy use. Much of that growth is expected to come in renewables (excluding liquid biofuels), which are forecast to grow from 8 percent of total U.S. energy use in 2011 to 11 percent by 2040. The energy provided by liquid biofuels is also projected to increase—from 1 percent of total energy use in 2011 to 2 percent by 2040—but the growth is expected to be far below the targets in the federal legislation, Gruenspecht noted, in part because of the issues related to cellulosic ethanol.
FIGURE 1-3 Sources of U.S. energy, projected to 2040.
NOTE: Because of rounding, percentages may not add to 100 percent.
SOURCE: EIA, 2013.
The percentage of total energy consumption provided by fossil fuels is projected to decline somewhat, but it will remain quite significant, which has implications for climate change, he noted. And the energy provided by oil and other liquids is projected to decline both in percentage and in absolute terms even though the growth in liquid biofuels is small. “I think that’s an interesting observation—there are other alternatives to oil beside biofuels.”
In the transportation sector, despite continued growth in travel, energy consumption by light-duty vehicles is projected to decline significantly. The consumption of motor gasoline is predicted to fall sharply during the period, dropping from 60 percent to only 47 percent of all transportation fuel. Meanwhile, diesel and liquid natural gas are expected to provide increasing percentages of transportation energy. This shift in the overall fuel mix arises from an effort to increase efficiency and thus decrease dependence on oil. Domestic oil production is projected to increase sharply in the next few years, reaching a peak around 2017 to 2020, after which it slowly decreases. The increase is due predominantly to the dramatically growing production of tight oil, or shale oil.
Because of the increase in domestic oil production, the import of liquid fuels—both petroleum and biofuels—is projected to continue dropping for the next several years and then remain relatively constant through 2040. “We don’t expect the reliance on net petroleum and biofuels imports to rise to where it was in 2005,” Gruenspecht said. “We see it remaining well below 40 percent [of total liquid fuels] throughout the entire forecast period.” There are also scenarios in which a more robust domestic production outlook combined with additional improvements in fuel efficiency, additional switching to alternate fuels, and various transportation applications led to a much smaller U.S. reliance on petroleum imports. “A situation in which the United States uses no net imported liquid fuels is a stretch,” he said, “but it’s certainly not beyond the realm of possibility.”
The projections for biofuels use show a relatively slow rise over the next decade. Total biofuels use in 2011 was just less than 15 billion ethanol-equivalent gallons; that is predicted to increase to about 19 billion gallons by 2022. This is far below the 36 billion gallons called for in federal legislation. Similarly, the cellulosic biofuels are projected to fall far short of the 16 billion gallons called for in the federal legislation; according to the projections far less than 1 billion gallons of cellulosic biofuels will be used in 2022. However, Gruenspecht said, “As the price
of oil goes up … and as technology improves, we do expect more biofuels to come in.” According to the projections, by 2040 more than 25 billion ethanol-equivalent gallons of biofuels will be used, including some 9 billion gallons of cellulosic biofuels.
In closing, Gruenspecht said that it is important to keep in mind that the market for motor fuels is very complex and that biofuels can play several different roles in that market. In particular, ethanol has played three distinct roles. It has been used as a source of octane, notably as a replacement for MTBE in reformulated gasoline following the phaseout of that additive. After MTBE was phased out, gasoline producers would pay almost anything for ethanol, he said. “You couldn’t sell fuel unless you had enough octane in the fuel, so it almost didn’t matter what it cost—it [the ethanol] was a small percentage of the fuel you had to buy.” A second role has been as a “volume enhancer,” Gruenspecht said. “It’s like a meat filler in the sense that you put in this stuff that has lower energy content per unit of volume, but people don’t notice.” Finally, ethanol also competes as an energy content provider. This is a more difficult market for ethanol, he noted, because ethanol costs more than gasoline per unit of energy.
At the moment, he said, ethanol is facing some significant challenges in moving beyond its current roles as a source of octane and as a volume enhancer. One challenge, Gruenspecht said is the “blend wall”—the difficulty in moving beyond 10 percent of overall gasoline sales. A second is the poor availability of E85 (gasoline fuel made with 85 percent ethanol) and other gasoline blends with a high percentage of ethanol. A third challenge is the difficulty of pricing E85 and other high-percentage blends to be competitive on an energy content basis.
Finally, Gruenspecht noted that the use of biofuels intersects with a number of public policy issues. First, the use of biofuels affects dependence on petroleum imports. It also can play a role in efforts to mitigate greenhouse gases. And it can have implications for rural economic development as well as food, water, environmental, and health policies. All of these should be considered in setting biofuels policies.
The session’s last speaker was Roger Prince, a senior research associate with ExxonMobil Biomedical Sciences in New Jersey. He offered the industry perspective on the production and use of biofuels.
Trends in World Energy Use
Prince began with a broad look at expected energy demand over the next several decades (ExxonMobil Corporation, 2013). The population of the countries of the Organisation for Economic Co-operation and Development is expected to stay more or less constant, while its gross domestic product (GDP) is expected to continue increasing steadily. However, although an increasing GDP has historically coincided with increasing energy use, energy efficiencies are expected to result in energy demand staying relatively flat in the next 30 years or more. By contrast, the rest of the world is expected to grow substantially in population, global energy demand, and energy consumption. Still, Prince said, there will be huge amounts of energy saved over what would normally have been expected because of the efficiencies that are coming into the market.
To meet that energy demand, the world will use a diverse mix of fuels that will change over time. Prince showed a figure illustrating the projected sources of the world’s energy from 2010 to 2040 (see Figure 1-4).
FIGURE 1-4 Worldwide fuel mix, 2010–2040.
SOURCE: Modified from ExxonMobil Corporation, 2013. Reprinted with permission from the ExxonMobil Corporation.
According to the projections, the world’s rising demand for energy will be met by increases in every category of fuel except for coal, which is forecast to decrease in total energy supplied toward the end of the period. The greatest increases will likely come from natural gas and renewables.
By far the largest amount of renewable energy now comes from biomass, Prince said. “But this is far from benign; this is the biomass that people use for cooking in the form of, say, animal dung or very green vegetation.” Because of the negative effects of burning this sort of biomass, governments and nongovernmental organizations are working to persuade people to stop using it, which should lead over time to a gradual slowdown in the increasing use of this material and, eventually, a decline, but that decline is not predicted to begin before 2030.
Renewables make up a relatively small part of the overall energy mix, and the category of “wind, solar, and biofuels” makes up a very small part of the renewable energy mix. Thus, despite recent efforts to increase the amount of renewable energy produced, Prince said, renewables still make up only a tiny sliver of the world’s overall energy budget. That percentage will grow in the next several decades but will still remain a relatively small part of the entire fuel mix.
One thing that is not widely appreciated, Prince said, is just how much energy people use on an individual basis—and how that individual use varies from region to region around the world. A human consumes about 2,000 kilocalories per day in food, he said, which is about 8,000 BTU. “If you think back to 1800, that was what the average person had to work with. If they went to chop wood, they were using that 8,000 BTU as an investment to chop wood to get wood.”
Today, by contrast, people in North America use an average of 740,000 BTU per person each day. This staggering increase in energy consumption is what allows our modern lifestyle. About 60 percent of that daily energy usage is indirect—used to make the various items that people use in their lives, from smart phones and food to roads and buildings—and about half of individuals’ direct energy usages is for personal vehicles. Energy usage per capita in North America is far higher than it is in any other region of the world—the closest regions are the Russian/Caspian region and Europe, both of which have average individual energy use of around 400,000 BTU—and the percentage of energy use devoted to personal vehicles is several times higher in North America than in any other region. Average energy use per person is less
than 150,000 BTU in Latin America and less than 100,000 BTU in Africa.
Thus, it is no surprise that energy use worldwide is expected to grow. “Approximately a quarter of the world’s population doesn’t yet have electricity,” Price said, “and something like a third of them don’t have modern cooking and heating. And those people are all going to want to have increased energy use.”
One problem with this increasing demand for energy is that it will come into conflict with the desire to keep greenhouse gases in check. A study carried out at the Massachusetts Institute of Technology examined what it would take to keep the level of carbon dioxide in the atmosphere from going above 550 parts per million, which is already significantly higher than the current level of 400 parts per million. Because the world’s population and the individual energy use per capita are both increasing steadily, various changes will be required to bring fossil fuel use down to a level that will prevent atmospheric greenhouse gases from continuing to climb. The study assumed that sharp increases in energy efficiency and decreases in demand per capita can be achieved in the next 70 years, but that still left a rapidly growing need for energy produced with little or no greenhouse gas emissions, such as nuclear power, fossil fuel plants with carbon capture technology, and renewables.
There is tremendous public support for bioenergy to be a large part of that renewable energy segment, Prince said. However, he cautioned, “I’m afraid it’s rather more optimistic than our view.”
To understand the earth’s overall capacity to produce bioenergy, it is useful to think in terms of its net primary productivity (NPP), which is “the amount of photosynthetic biomass available for exploitation by the biosphere.” It can also be thought of as the total amount of carbon dioxide taken in by plants during photosynthesis minus the total amount of carbon dioxide released by plants through respiration. The earth’s NPP is staggeringly large, Prince said—about 56 gigatonnes of carbon dioxide per year.
It turns out, he said, that humans already use about 30 percent of that NPP for food, fabrics, construction, and other uses. Any increase in the use of biofuels will require an increase in the amount of the earth’s NPP being appropriated for human use. And of course, Prince noted, essentially 100 percent of this NPP is already being used—if not by
humans, then by various animals, fungi, bacteria, and other forms of life—so any increase in the human use of the NPP will take away from other ecosystem processes. One possibility would be to increase the earth’s total NPP—to grow crops, such as algae, where there is currently little growing. “But we already do an amazingly good job at harvesting the natural world,” he said, “and it will be very hard for us to harvest a lot more without displacing every orangutan and every zebra.”
Given these constraints, it is clear that a crucial factor in the success of biofuels will be how efficiently they can be produced. And unfortunately, the first-generation biofuels that are in use today—ethanol from corn, sugarcane, wheat, sugar beets, and other plants, and biodiesel from vegetable oils from such plants as soybeans and rapeseed are simply not very efficient in harvesting energy from the sun. Calculating efficiency in terms of how much energy is available from the biofuel versus how much energy was in the sunlight hitting the plants used to produce the biofuel, the efficiency of ethanol made from corn is only about 0.03 percent. Ethanol made from sugarcane is much more efficient but still only about 0.14 percent (Kheshgi et al., 2000). So, there are serious questions, Prince said, as to whether biofuels can be efficient enough to contribute significantly to the world energy budget.
“It is generally agreed that in the United States we make about a 30 percent energy profit when we make ethanol,” Prince said, “so that means that you need to produce four liters of ethanol to export one from the farm-distillery complex to have a self-contained energy system.” This implies, among other things, that it would be essentially impossible to use ethanol to completely replace gasoline, he said. “If you wanted to be entirely petroleum free, you’d have to produce 4 gallons of ethanol for every gallon you exported to the consumer. And that makes it an impossibility at current scales.”
There are other factors that must be taken into account as well in discussing ethanol’s ultimate potential. One important factor is that corn production causes substantial soil loss. “The Midwest loses about 4 kilograms of soil for every liter of ethanol that’s made in the Midwest,” Prince said. “It all goes out down the Mississippi. It’s an interesting question as to how long one can continue doing that.” There are efforts under way to use no-till farming to reduce the soil loss, but their widespread adoption is still uncertain.
A positive factor is that the major byproducts of converting corn into ethanol are used as animal feed. “You get a roughly equal weight of ethanol and animal feed when you convert corn into ethanol,” Prince
said. “That’s very important for the economics. It’s also important in that dried distillers grains can’t be stored very long, so they have to be put through the animals quite quickly, and that keeps the price of meat gratifyingly low.”
Producing ethanol from sugarcane yields no byproduct that can be used as animal feed, Prince noted. “The rest of the plant is burned. There are no real co-products other than electricity.”
Another byproduct of ethanol production is carbon dioxide. Today, Prince said, the carbon dioxide is generally captured and sold for use in carbonating beverages, but it could also be used in carbon sequestration. Because the gases from a fermenter are pure carbon dioxide, it is straightforward to compress them and bury them underground, and, indeed, there is a pilot plant at the Arthur Daniels Midland distillery in North Dakota that is doing just that. In Brazil, the carbon dioxide production from distilling ethanol from sugarcane is so efficient that if the carbon dioxide byproduct were to be captured and put underground, the net result of producing and burning the ethanol would be to remove carbon dioxide from the atmosphere (Kheshgi and Prince, 2005). A pilot project has proven the technical feasibility of the process, but it remains to be seen whether it can be scaled up economically to a degree that will make a significant contribution to greenhouse gas reduction.
The efficiencies for biodiesel are different and even less promising than for ethanol. The biodiesel yield is generally about 50 to 60 gallons per acre of planted crop, as opposed to 250 gallons per acre for ethanol. As with ethanol production from corn, a byproduct of biodiesel production from soybeans can be used as animal feed, and it plays an important role in the economics of biodiesel production. And as with ethanol production, there are serious problems of soil erosion that accompany the production of biodiesel.
Other Bioenergy Pathways
There are many biofuels other than ethanol that can be made from plants, Prince noted. The sugars from corn or sugarcane can be transformed through fermentation to not only ethanol but also acetone, butanol, and isobutanol, among others. The fermentations are done anaerobically using either bacteria or yeast. There are several companies trying to commercialize sugar-derived butanol right now, he said.
There are also aerobic systems where algae and yeasts convert sugars into oils. “These are important, for example, in baby formula,” Prince
said. It is also possible that engineered yeasts can be used to make such molecules as farnesene and farnesane, which are branched small alkanes.
The challenge with these approaches is that they all require sugar that is low-cost and of reasonable purity. Corn syrup works well, but it is expensive, and the yields are small enough that it is not yet commercially viable. Several companies are moving to Brazil to use sugarcane as the raw material for such processes.
Eventually, developing commercially viable biofuels will require much cheaper and more abundant sources of glucose than corn or sugarcane, Prince said. Thus, it would be valuable to find an efficient way to convert cellulose into sugars that can then be turned into these products. The cellulose could come from crop residues or specially grown crops. But finding an economical way to use cellulose has proved challenging.
One problem is simply crop supply: The trees and grasses that are the most efficient producers of cellulose are not yet farmed, and the crop densities preclude long-distance shipping. Another problem, Prince said, is that although it is reasonably easy to grow cellulose, separating the cellulose from the lignin that comes with it in the plant is a challenge, and it seems to be a different challenge for every different plant. Furthermore, separation processes tend to generate microbial inhibitors.
One of the biggest challenges is the economics of the system, Prince said. “Many predictions imagine that farmers will be willing to sell cellulosic biomass for the order of $40 or $50 a ton. Today, they can get $200 a ton for hay.” Given the difference, it is unlikely that many farmers will switch over from hay to cellulosic biomass.
Looking ahead, Prince said that there are a number of bioenergy pathways emerging—ways in which bioenergy could contribute to the overall energy supply. A simple one that is already being used in Scandinavia is burning biomass to make electricity and distributed home heating. Anaerobic conversions can be used in various ways, such as to first gasify the biomass and then convert the resulting gases—e.g., methane and carbon monoxide—by a microbial process into liquid fuels. “People have started working on that,” he said, “but right now the price of natural gas is so low that they’re moving from using biogas to moving back to natural gas.”
Gas from the gasification of biomass could be turned into methanol which could then be converted into gasoline. This can be done on an economic scale, Prince said, but it is not practiced today.
As previously described, cellulosic biomass can be treated to separate out the lignin from the cellulose, which can then be fermented to produce ethanol and perhaps other biofuels such as butanol and other alcohols.
Further out on the horizon is the growing of algae for use in producing biofuels. “The nice thing about algae,” Prince said, “is that they grow all year round, and you can imagine that you could adjust their concentration so they’re very efficient users of sunlight.” On the other hand, one of the challenges will be “to persuade the algae to make oil rather than to make more algae.”
In conclusion, he said, the main question about biofuels is whether they can be produced on a large enough scale to be of any real importance to the world. The world already uses an enormous amount of energy—the equivalent of more than 250 million barrels of oil per day—and that number will certainly increase. “These numbers are just so staggering that it’s a real challenge to imagine biofuels having a significant impact,” Prince said. “And as you do produce them, can you do it with an acceptable environmental impact? As I’ve said, naysayers could compare current biofuels with mining—we lose more soil than we gain in fuel from current farming practices. So, in the long run, whether we can do this is a real challenge, and I don’t think the answers are at all clear yet.”
To start off the discussion following the presentations, session moderator Lynn Goldman asked Gruenspecht if all of the increased use of ethanol in fuel has been motivated by mandates, or if some of it has been market-driven. The sales of ethanol as an octane enhancer were certainly market-driven, he said. “I guess the question is what drove MTBE out and how legitimate those concerns were,” but once MTBE was no longer available, there was a real need for octane enhancement, and ethanol filled that need. The use of ethanol as a volume enhancer in fuel has not been driven by the mandates, either, he said, but it has been driven by the existence of the tax credits for producing ethanol from corn. “People were looking for the cheapest volumes, and with the tax credits on a volume basis the ethanol was cheaper, so retailers and distributors were happy to put as much of ethanol into the fuel as they could, which was that 10 percent blend wall,” Gruenspecht said.
Carlos Santos-Burgoa from the Pan American Health Organization asked whether there is any information on the actual impact of ethanol production on the cost of corn for consumers, especially in countries where corn is a major source of food. Gruenspecht answered that studies in the literature come up with a wide range of numbers, so that there is no well-supported answer to the question. However, he said, he suspects that “this is an area where the community has possibly been affected by the fact that there are interests at stake.” Such a situation should not influence the science, he said, “but I think it does.”
Santos-Burgoa also asked about the issue of soil erosion and other potential impacts in countries where forests are being clear cut to provide land for growing crops to produce biofuels. Gruenspecht answered that converting any biome to a farming biome has “a huge energy cost, a huge cost in soil carbon, a huge cost in erosion, all those things.” The same thing is true for cutting down rainforest to grow sugarcane as for cutting down deciduous forest to put in a shopping mall. “There’s a huge cost environmentally, and it needs to be taken into account. Usually, we don’t bother to worry about it. We should,” Gruenspecht said.
Catherine Kling from Iowa State University commented that it is not clear that all of the food prices seen around the world can be attributed to the biofuels mandates. The literature clearly indicates that biofuels production has had some effect, “but there also have been some important droughts, and there have been some very bad policies that foreign governments have undertaken in response to price increases, such closing markets, which actually aggravated it.”
More broadly, she said, it is important to look carefully at the connection between ethanol production and food prices and also the connection between food prices and health. First, she noted, “Corn went from $4 in 2007 to about $7 now. That cannot translate into a tripling of food prices. So I’m not quite sure where that number comes from, but I’d just be a little careful about some of those world prices.”
Second, it is not clear that higher prices for corn are always a bad thing. In the United States, for instance, higher corn prices lead to higher prices for meat and corn sweetener, and if consumption of those two items drops it is probably good for health. “It’s a very different question if you’re looking at food prices internationally,” Kling continued. “There people don’t have heavy meat diets. There are places in the world where people are definitely starving and these food prices changes can be very significant.” Even so, she said, it is important to ask whether the right way to address that particular issue is to tinker with biofuels or whether it
would make more sense to provide more food aid, technology, lump-sum payments, or other ways to improve food security.
Panel member Al McGartland asked about the issue of transporting biofuels. “As I read some of the literature,” he said, “ethanol is largely transported by truck and not by pipeline. And I’m wondering if we’re going to expand biofuels, would that be an obstacle?” Gruenspecht explained that most ethanol is transported by rail or by truck. It is difficult to move through pipelines because of the presence of water in the pipelines. One option would be to use the second-generation biofuel biobutanol in place of ethanol, as it does not have the same problem with exposure to water. Biobutanol has a higher energy density than ethanol, but does not provide the same octane enhancement, Gruenspecht said. Other biofuels are also under consideration, but they have other issues. In short, there is no silver bullet that will solve the transport issue for biofuels. “There’s only silver buckshot, so we should be doing everything.” However, if there are a number of different types of biofuels, each with its own transportation infrastructures, that would become a constraint in and of itself, he said. “So, I think there are some real challenges related to those issues. But it’s just kind of hard to get into.”
EIA (U.S. Energy Information Administration). 2013. Annual energy outlook 2013: With projections to 2040. Washington, DC: U.S. Energy Information Administration, Department of Energy. Available at http://www.eia.gov/forecasts/aeo (accessed June 30, 2013).
ExxonMobil Corporation. 2013. 2013: The outlook for energy. A view to 2040. Irving, TX: ExxonMobil Corporation. Available at http://www.exxonmobil.com/Corporate/Files/news_pub_eo2013.pdf (accessed July 28, 2013).
HLPE (High Level Panel of Experts). 2013. Biofuels and food security. A report by the High Level Panel of Experts on Food Security and Nutrition of the Committee on World Food Security, Rome 2013. Available at http://www.fao.org/fileadmin/user_upload/hlpe/hlpe_documents/HLPE_Reports/HLPE-Report-5_Biofuels_and_food_security.pdf (accessed July 30, 2013).
Kheshgi, H. S., and R. C. Prince. 2005. Sequestration of fermentation CO2 from ethanol production. Energy 30:1865–1871.
Kheshgi, H. S., R. C. Prince, and G. Marland. 2000. The potential of biomass fuels in the context of global climate change: Focus on transportation fuels. Annual Review of Energy and the Environment 25:199–244.