The Ethanol Answer to Carbon Emissions
When the United States gets serious about the threat of global climate change, it should turn to ethanol to power cars.
The moment is fast approaching when the United States will have to face up to the need to reduce greenhouse gas emissions. The Intergovernmental Panel on Climate Change is finding growing scientific evidence that human activities are forcing a gradual warming of the planet, and recent international negotiations in Kyoto, Bonn, and Marrakech have demonstrated that the world’s political leaders are taking the threat more seriously. The United States opted out of the most recent rounds of negotiations, and many analysts have pointed out weaknesses in the Kyoto Protocol. Although there are large problems with the Kyoto agreement and its focus on unrealistic near-term targets for U.S. emissions reductions, the United States cannot ignore the need to reduce carbon emissions over the long term. The November 2001 Marrakech meeting concluded that the United States and other developed nations should reduce their emissions to 95 percent of the 1990 level in approximately 10 years. Although the United States rejected that goal, eventually it will have to reduce its emissions even further.
The United States is responsible for a quarter of the world’s total carbon emissions, and Americans’ per capita emissions are five times the world average. A major source of carbon emissions is the U.S. personal transportation system. Light-duty vehicles–cars, sport utility vehicles (SUVs), minivans, and other light trucks–are prolific CO2 emitters, producing 20 percent of total U.S. emissions.
The fuel economy of the average new vehicle has been declining because of the increasing market share of SUVs and other light trucks. Since 1990, gasoline consumption (which is proportional to vehicle CO2 emissions) has increased 19 percent because of the change in vehicle mix, an increase in the number of vehicles, and increases in vehicle miles traveled. If these trends continue, major reductions in CO2 emissions from cars and light trucks will be impossible; even attaining the Marrakech goals will be difficult.
A recent National Research Council (NRC) committee concluded that fuel economy could be improved by 55 percent by a series of small improvements, without changing vehicle size or performance; the increased manufacturing costs would be offset by the fuel saving over the lifetime of the vehicle. However, the committee found that phasing in these more fuel-efficient vehicles would be counterbalanced by the increase in fleet size and the shift toward larger, more powerful vehicles. Thus, taking advantage of the technologically and economically feasible efficiency options would result in little or no reduction in gasoline consumption.
Americans are not unique in wanting large vehicles with powerful engines. High-income consumers in Europe and Japan also want these vehicles, despite gasoline prices of more than $4 per gallon and high taxes on fuel-hungry vehicles. Comparing European and U.S. fuel prices and average fuel economies, it appears that roughly tripling the price of fuel is associated with about a 30 percent increase in fuel economy. Assuming that the demand response is proportional to price, this experience suggests that a further tripling of price would be required to induce drivers to choose a vehicle mix that averaged 50 miles per gallon. We doubt that a democratic government would be able to increase fuel taxes to a level that would raise the price of gasoline to the $13 per gallon range. Since the early 1980s, low U.S. fuel prices have induced Americans to ignore efficiency in their vehicle choices. The federally mandated Corporate Average Fuel Economy (CAFE) standards have forced automakers to offer more small and fuel-efficient cars, but consumers have flocked to SUVs, minivans, and light trucks that have a less stringent CAFE standard. With these larger vehicles now accounting for more than half of new vehicle sales, the fuel economy of the average new vehicle has been declining. Even if we could reverse this trend, the NRC study indicates that it would still be difficult to achieve a substantial reduction in total gasoline use and carbon emissions. The only practical path to achieving significant emissions reductions is to find an alternative to gasoline as a fuel.
Three technologies have the potential to power motor vehicles with no net CO2 emissions. The first is batteries, where the electricity to charge the batteries comes from renewable energy or nuclear power. The second is fuel cells, where the hydrogen is produced by renewable energy or nuclear power. The third is an internal combustion engine using ethanol from cellulosic biomass that is grown and processed with no fossil fuels.
Battery-powered cars are expensive and a potential public health menace. To get even a 100-mile range, about 1,100 pounds of batteries are required for a two-passenger car. Making and recycling these batteries is expensive, leading to large increases in the cost of driving. Mining and smelting the heavy metals for the batteries, as well as making and recycling the batteries, would discharge large quantities of heavy metals into the air, water, and landfills. If the current U.S. fleet of 200 million vehicles were run on current lead acid, nickel cadmium, or nickel metal hydride batteries, the amount of these metals discharged to the environment would increase by a factor of 20 to 1,000, raising vast public health concerns. Unless there are major breakthroughs in electrochemistry, this is not an attractive strategy.
A great deal of attention has gone to fuel cells, which emit nothing but water vapor and are much more efficient than an internal combustion engine. Unfortunately, the current reality is that fuel cells are extremely expensive, and they cannot match the driving performance of current engines. Major technological breakthroughs are required to make fuel cells attractive for light-duty vehicles. The environmental implications of fuel cells cannot be known until we know what materials and processes will be used and how the hydrogen will be produced.
The technology that is most attractive and available is ethanol made from grasses and trees. Because 95 percent of the ethanol currently produced in the United States is made from corn, it is critical to understand how this process differs from what we are recommending. Corn is a very small part of the corn plant and contains less than half the total energy. Although it is easier to process the corn than the rest of the plant, the ethanol yield is small. The net energy from producing ethanol from corn is perhaps only 25 percent of the energy in the ethanol, with most of the energy used in processing the ethanol coming from petroleum and natural gas. No one who set out to produce ethanol would first grow food. Only someone who was interested primarily in subsidizing U.S. farmers would consider corn as a source of energy.
The amount of biomass, and thus the amount of ethanol, that can be produced per acre of land is much greater than the amount of corn that can be produced. Furthermore, low-grade land unsuitable for producing corn can produce biomass. Corn requires fertilizer and pesticides, and often irrigation as well; other forms of biomass can thrive without these inputs. Current farming practices result in large soil losses; growing biomass would essentially end soil loss, because once the first crop was planted, the soil would almost never be uncovered. If the biomass itself is used to power the production process and make any fertilizer, no fossil fuels need be used. Net energy would be about 75 percent of the gross energy produced.
U.S. ethanol production from corn is cost-effective now because of a $0.55 per gallon tax subsidy and a high value for byproducts from the process. Dry milling of corn produces an animal feed supplement as a byproduct, and wet milling produces starch sweeteners, gluten feed and meal, and corn oil as byproducts. Sale of these byproducts is an important source of income, but the market for these byproducts is likely to be saturated once total annual ethanol production reaches 5 billion gallons. When the byproducts are no longer valuable, cellulosic ethanol is predicted to be cheaper to produce than ethanol from corn.
Ethanol production from lignocellulosic feedstocks is undergoing rapid development with prototype pilot and full-scale plants under development. The main difference between corn processing and using cellulosic feedstocks is that the fermentable sugars in cellulose are more tightly bound. Freeing the sugars to permit fermentation requires more intense processing than is needed to remove starch from corn. However, an advantage of cellulose processing is that it also yields lignin that can be burned to provide energy to run the process and to generate electricity that can be sold.
Growing corn, wheat, rice, and sugarcane produces large amounts of agricultural wastes, some of which are burned, degrading air quality. In the production of cellulosic ethanol, the bulk of the biomass would become a valuable source of energy rather than a waste product. In fact, municipal solid waste (MSW) includes a large volume of cellulosic material that has the potential to be converted to ethanol. Because this material would have to be separated from other parts of the MSW, it is more expensive than energy crops. However, cities in the Northeast such as New York and Philadelphia have paid as much as $150 per ton to dispose of their MSW. At this price, it would be worth sorting the MSW to remove the cellulose and other materials.
To grow enough biomass to enable ethanol to replace gasoline would require an enormous amount of land. To provide sufficient ethanol to replace all of the 130 billion gallons of gasoline used in the light-duty fleet, we estimate that it would be necessary to process the biomass growing on 300 million to 500 million acres, which is in the neighborhood of one-fourth of the 1.8-billion acre land area of the lower 48 states. Most U.S. land is now grassland pasture and range (590 million acres), forest (650 million acres), or cropland (460 millions acres). The remaining acreage is used for human infrastructure, parks and wildlife areas, and marsh and wetlands. The 300 million to 500 million acres could be supplied from high-productivity land (39 million acres of idled cropland), from land currently used to grow grain that is sold below production cost (approximately 45 million acres), and from pasture and forestland that are not associated with farms. No land from national parks, wilderness areas, or land for buildings, highways, or other direct human use would be required.
If the goal is to minimize environmental intrusion, the emphasis should be on using a very large land area with a diverse array of trees and grasses that would not have to be harvested annually. With current technology, trees would be harvested by means of common forestry practices and grasses would be cut and baled like hay. Research is being done and should continue to develop harvesting practices that are most economically effective and environmentally benign for reaping biomass feedstock for ethanol production.
The United States is not the only country where ethanol production would make sense. Nations such as Brazil, Argentina, and Canada could grow enough biomass to produce ethanol for their own needs and for export. Indeed, Brazil has long produced ethanol as an auto fuel. Biomass ethanol could fuel a major proportion of the world’s automobiles, leading to a considerable reduction in CO2 emissions, making resource use more sustainable and reducing soil loss.
Although the United States could produce sufficient ethanol from energy crops to run all its cars and light trucks, this does not mean that it necessarily should do so. We need to look closely at the economic and environmental considerations.
The economic case against making a massive commitment to ethanol is that cellulosic ethanol currently costs too much to compete with gasoline. The refinery gate price of gasoline is about $0.80 per gallon; transportation, storage, and retailing add about $0.40 per gallon; and taxes raise the price at the pump to roughly $1.50 per gallon. Producing cellulosic ethanol costs about $1.20 per gallon (1.80 per gallon, gasoline equivalent, since ethanol has two-thirds of the energy of a gallon of gasoline). Assuming that the per-gallon distribution costs are the same for ethanol and holding total tax revenue constant, ethanol would sell for $1.80 per gallon at the pump. However, this is equivalent to $2.70 per gallon in order to get as much energy as in a gallon of gasoline. Technology improvements promise to reduce this cost, but it is unlikely to fall below the cost of producing gasoline.
Motorists will not switch to ethanol unless the price of gasoline is at least $2.70 per gallon. Although this price will seem astronomical to U.S. drivers, it is actually much lower than the price that would be required to convince Americans to buy 50-mile per gallon cars or the price that the Europeans and Japanese are paying now. If the United States decides that the motor vehicle sector must reduce its carbon emissions, it will be much easier to convince them to switch to ethanol fuel, even at $2.70 per gallon, than to convince them to drive smaller cars or cars with smaller engines.
Growing its own motor vehicle fuel would also make the United States less dependent on imported oil, which will enhance its political independence. It would be a gift to future generations to bequeath to them a stable fuel supply that is not subject to wars, civil unrest, or global politics. Farmers would be major beneficiaries. They could stop growing grains that fetch a price that is lower than the production cost. Growing energy crops would generate more than $100 billion in revenue to farmers and more than $80 billion in revenue to ethanol producers located within 30 miles of where the energy crops are grown. Since workers employed in hauling the energy crops to the ethanol plant and those working in the plant would be rural workers, this program would contribute $100 billion to $180 billion to rural and farm revenue.
Potentially more formidable than the economic barriers to ethanol would be the public reaction to using roughly a quarter of U.S. land for energy crops. Using land that is not currently being cultivated for crops is likely to raise the hackles of environmentalists and hunters, who will argue that the “natural” ecology is being destroyed. However, this milk has already been spilled. Little U.S. land has been spared from human activity. Almost all of the 460 million acres of cropland and the 590 million acres of grassland pasture and range have been altered from their native state as wetlands, forest, grassland, or other natural ecosystems.
The principal energy crops would be grasses such as switchgrass, which is a native prairie grass, and hybrid trees such as poplars or willows. A well-planned and thoughtful bioethanol program could return much of that land closer to its native state, enhancing the environment, as well as bringing the benefits of a renewable and sustainable fuel. Properly managed, the energy crops could help endangered species and enhance recreational opportunities. This proposal amounts to restoring much of the Great Plains to tall grasses. To be sure, the grass would be mowed annually, but there would still be plenty left to feed roaming bison, deer, and elk. Certainly, these grasslands and forests would create habitats for birds and other creatures, as well as land for hiking and other recreation. Providing environmental benefits such as these will be essential to making this fundamental shift in land use politically palatable.
Minimizing the land required to replace gasoline requires dense plantings of energy crops. Combinations of other plant species (whether for energy or not) would provide species diversity, encouraging animal diversity. Native plants have already demonstrated that they can thrive without human inputs such as water and fertilizer. Strategic placement of plantings could provide habitat connectivity to increase a pressured species range. It would require more land than would a strategy that emphasized planting just the most energy-rich species, but that might be a tradeoff that the U.S. public would be willing to make.
Staging the transition
If the United States decides to make the switch from gasoline to cellulosic ethanol, implementing that decision will pose formidable difficulties. The most appealing fuel appears to be E85, a mixture of 85 percent ethanol and 15 percent gasoline. Pure ethanol has an extremely low vapor pressure, which would make it difficult to start the engine on cold days. Adding a small amount of gasoline would overcome the problem.
E85 cannot be used in most of the cars on the road. Changes in the engine would be necessary. Congress already encourages the use of alternative fuels by giving a substantial fuel economy credit to flexibly fueled vehicles (FFVs). These light-duty vehicles, which cost about $250 more to manufacture, are capable of using gasoline-ethanol blends up to E85. The federal subsidies enable the manufacturers to sell FFVs for less than the cost of conventional vehicles. The attractive price has resulted in the sale of about four million FFVs (a combination of cars and light trucks), but almost none of them use E85 because the price is too high.
Before automakers would produce vehicles optimized for E85 and before customers would buy them, there would have to be a guarantee that there would be a substantial supply of this fuel universally available at an attractive price. Before large investments would be made in producing cellulosic ethanol, farmers and ethanol processors, distributors, and retailers would have to be assured that there will be a considerable demand for this fuel at a price that promises an attractive return. Thus, there is a chicken and egg problem: Which comes first, the investment in cellulosic ethanol or the investment in motor vehicles?
We think that increasing the supply of cellulosic ethanol should come first. All cars can use E10, a mixture of 10 percent ethanol and 90 percent gasoline. With a little modification, today’s vehicles could use fuel mixtures with up to 22 percent ethanol. Since the light-duty fleet uses 130 billion gases of gasoline annually, mandating that all gasoline be E10 would require 13 billion gallons of ethanol, roughly six times the current production of fuel ethanol. If E22 were mandated, 30 billion gallons of ethanol would be required. These are large levels of demand for a fledgling industry that produces almost no ethanol from cellulosic biomass at present.
Requiring that all gasoline be E10 is not feasible when current ethanol production is only 2 billion gallons. Regulators cannot simply order more ethanol to be added to gasoline. A more efficient way of increasing ethanol production would be a plan that increases the tax on gasoline in order to subsidize the production of ethanol. The price would be calculated to maintain the current government tax revenue and to subsidize ethanol by at least its cost premium of $0.80 per gallon ($1.20 per gallon of gasoline equivalent). The price structure could be set so that one would pay a lower price per gallon of gas equivalent the higher the percentage of ethanol in the fuel mix, up to E85. In that way, consumers would be likely to move immediately to E10, then to pay for modifications to their engines so that they could use E22, and eventually to buy FFV vehicles so that they could take advantage of the savings from using E85 fuel. The transition would occur over time, so that producers, distributors, and service stations would have time to scale up and make other adjustments as demand grew. The government subsidy could be reduced gradually as the cost of producing ethanol declines because of advances in technology and economies of scale. However, since ethanol is likely to continue to be more expensive than gasoline, the cost of E85 will be higher than the cost of straight gasoline, unless there are higher taxes on gasoline.
The dynamics of increasing cellulosic ethanol production are important. Since there are no commercial cellulosic ethanol plants now, there is a great deal to be learned. By increasing production gradually, we will be able to learn from experience and quickly incorporate insights into the design of new plants that will be coming online steadily. This deliberate approach slows the increase in production but lowers costs. At some stage, the process technology would be optimized and new plants would be built in parallel to increase capacity rapidly.
The combination of the federal subsidy for ethanol production and the growth in demand would be a magnet for R&D and infrastructure investment. It would not be hard to generate private-sector interest in what could become a $200 billion industry
One way of looking at the program is that in addition to achieving its primary goal of reducing greenhouse gas emissions, it would move resources used to protect our oil supplies to producing fuel at home, and it would eliminate the need for tens of billions of dollars a year in farm subsidies.
Looked at in strictly technocratic terms, this approach is compelling and far more reliable than the alternatives of switching to cars powered by electric motors or hydrogen-based fuel cells. However, two political contingencies overshadow any technological quibbles. The first is whether (and when) the United States will make a commitment to reduce greenhouse gas emissions. Without an agreement that the nation will commit extensive resources to achieving this goal, there is no chance that the public will support a massive subsidy for ethanol production. And unless the public is willing to consider an unprecedented change in the way that land is used, producing the biomass necessary for large-scale ethanol production would be impossible.
Although these contingencies are formidable, they shrink to manageable size beside the potential devastation that could accompany rapid global climate change. And although a transition to ethanol-based vehicles is a daunting challenge, it is actually less difficult to envision than a switch to completely different power system for vehicles.
Harron S. Kheshgi et al., “The Potential of Biomass Fuels in Context of Global Climate Change: Focus on Transportation Fuels” Annual Review of Energy and the Environment 25 (2000): 199–244.
Charles E. Wyman “Biomass Ethanol: Technical Progress, Opportunities, and Commercial Challenges,” Annual Review of Energy and the Environment 24 (1999): 189–226.
Lester B. Lave ([email protected]) is the James H. Higgins Professor of Economics and University Professor at the Graduate School of Industrial Administration at Carnegie Mellon University in Pittsburgh. W. Michael Griffin is the executive director of the Green Design Initiative at Carnegie Mellon. Heather MacLean is assistant professor of civil engineering at the University of Toronto.