Sixty billion gallons by 2030: economic and agricultural impacts of ethanol and biodiesel expansion.

Use of bioenergy feedstocks to produce transportation fuels could not only help reduce reliance on foreign oil, but could also provide significant environmental benefits and invigorate rural economies. Agriculture is well positioned as a feedstock source because the fuels can be utilized with current engine technologies and are compatible with the current distribution infrastructure. The anticipated commercialization of cellulose-to-ethanol technology will enable fuels to be derived from a diverse portfolio of feedstocks from numerous regions of the country.

The Energy Policy Act of 2005 establishes a renewable fuel requirement for the nation, mandating 7.5 billion gallons of renewable fuels by 2012. A more sweeping renewable fuels standard was proposed as part of The Biofuels Security Act of 2007 (sponsored by Senator Tom Harkin and co-sponsored by Senators Lugar, Biden, Dorgan, and Obama) (Harkin, et al., 2007). The Governors' Ethanol Coalition has recommended replacing at least 25% of petroleum used as transportation fuels by the year 2025 (Governor's Ethanol Coalition 2006).

The objective of this study is to project the U.S. agricultural sector and economic impacts of increasing ethanol and biodiesel production beyond the levels specified in the recently enacted renewable fuel standard. The levels of ethanol production analyzed are 10, 30, and 60 billion gallons of ethanol annually by 2010, 2020, and 2030, respectively. Sensitivity to the timing of commercialization of cellulose-to-ethanol technology and impacts of associated corn to ethanol industry adjustments is projected. Impacts of producing 1 billion gallons of biodiesel production by 2012 and 1.6 billion gallons by 2030 are also projected.

Prior Research

Several studies have addressed various aspects of bioenergy production and contribution toward renewable energy (USDA-OCE 2002; Urbanchuk 2001; Shapouri, Duffield, and Wang 2002; Sheehan 2002; Walsh et al. 2003; McLaughlin et al. 2002). Previous economic modeling evaluating agriculture feedstocks for energy has been conducted in the context of carbon displacement potential (McCarl and Schneider 2000). Adjustment costs incurred in the short run for implementing new technologies and/or policies are not considered by these models (Schneider 2000). The potential regional economic impacts of converting corn stover to ethanol have been projected using IMPLAN (English, Menard, and De La Torre Ugarte 2000).

Methodology

Key methodological steps to conducting the analysis include: definition of biofuels goals, selection of representative conversion technologies, collection of associated cost information, definition of key assumptions, update and expansion of POLYSYS (a dynamic agricultural sector model), development of a program to integrate POLYSYS results into IMPLAN (PII), modification of IMPLAN (economic input output model) to accommodate biomass feedstock production and biofuels conversion industries, and scenarios development (De La Torre Ugarte et al. 2006).

The conversion technologies include ethanol from shelled corn, cellulosic residues (stover, switchgrass, and wheat straw), food residues, wood residues (forest residues, mill wastes, fuel treatment and forestland thinnings), and biodiesel from soybeans and from yellow grease/tallow. Representative facility output, feedstock use, and associated costs are developed based on prior studies (McAloon et al. 2000; personal correspondence from V. Eidman 2006; Aden et al. 2002; BBI International 2002; English, Jensen, and Menard 2002; Fortenberry 2005). Conversion coefficients of cellulose to ethanol are increased linearly for stover, straw, and dedicated energy crops from 2015 to 2030. Conversion coefficients of feedstocks to corn grain ethanol and biodiesel are assumed to increase through 2019 and thereafter remain steady.

Several key study assumptions are required in addition to setting biofuels targets and selecting conversion technologies. Cellulose-to-ethanol technology is assumed to be commercially available in 2012. Switchgrass serves as a proxy for dedicated energy crops, with yields increasing over time that range from 1.5 to 5%, depending on region. No-till adoption increases from 20 to 55%. The available land base includes 307 million acres of cropland in major crops plus hay and 56.2 million acres of cropland in pasture. Maximum percents of Distiller's Dry Grains (DDG's) in feed rations are assumed at 30% for beef and 10% for dairy, hogs, and broilers.

The targeted biofuels production levels plus data on conversion costs for agricultural and forest feedstocks are introduced into POLYSYS to estimate the quantity and type of energy to be produced from agriculture, as well as the price, net farm income, and other agricultural sector impacts (De La Torre Ugarte and Ray 2000). In regions where dedicated energy crops are determined to be profitable, pastureland has the potential to be converted to dedicated energy and other crops. Loss of forage production is replaced with new hay production. The livestock module in POLYSYS is an integrated version of the Economic Research Service's econometric livestock model. To project the potential of dedicated energy crops to provide feedstocks, enterprise budgets and yields for switchgrass are incorporated into POLYSYS. Production is assumed to be suitable on 368 million acres.

To evaluate the potential of crop residues to provide feedstocks to the bioproduct markets, POLYSYS includes corn stover and wheat straw response curves that estimate stover and straw quantities as a function of corn and wheat grain yields, as well as stover and straw production costs as a function of yields of removable residue. Estimated response curves are obtained through the Oak Ridge National Laboratory (Walsh et al. 2003). Residues needed to keep erosion at less than or equal to the tolerable soil loss level are incorporated. Total quantities of corn stover and wheat straw that can be collected in each county are estimated for each tillage and dominant crop rotation scenario. The costs of collecting corn stover and wheat straw include baling, staging, and replacing nutrients.

The cost of transporting biomass feedstocks from the farm gate to the production facilities is added to conversion costs. An iterative process is used until the corresponding price of biomass is equal to current iteration biomass price. Once the iterative process has converged and equivalent costs of ethanol production exist, the model has determined the optimal market level of feedstock quantities. The price at which these wood residues feedstocks come into use is determined by regional harvesting costs plus transportation costs. Distiller's dry grains from ethanol production and soybean meal from biodiesel production are integrated within the model to evaluate how their quantities and prices affect the final market equilibrium.

For biodiesel, beef and poultry wastes are modeled as a function of beef and poultry cash receipts, respectively. Yellow grease from food waste is a function of population. Soybean meal byproduct from crushing enters into the POLYSYS soybean product module where prices are endogenously determined.

An interface program, the POLYSYS/ IMPLAN Integrator (PII), developed at The University of Tennessee, takes POLYSYS projections of acreage, price, change in government programs, and cost output and automates changes to IMPLAN data bases (English, Menard, Wilson, and De La Torre Ugarte 2004). PII adds an energy crop sector to IMPLAN based on information supplied by POLYSYS. A renewable energy sector is added to each state's IMPLAN model, and the operating impacts from the renewable energy sector are estimated. IMPLAN employs a regional social accounting system to generate a set of balanced economic/social accounts and multipliers (MIG 1999). The model estimates total industry output, employment, and value added for over 500 industries.

Results under three scenarios are compared with a baseline scenario called USDAExt, which is an extension of the 2006 USDA baseline. The first scenario projects the impacts of attaining the targets assuming the cellulose-to-ethanol technology would be commercially available by 2012 (ETH60). In this scenario, use of corn grain is kept at near production capacity of plants, even with introduction of cellulose-to-ethanol technology. The second scenario allows the corn grain-to-ethanol industry to adjust as cellulose-to-ethanol technology becomes commercially available in 2012 (ETH60CA). In the third scenario (ETH60CACD), the cellulose-to-ethanol technology would be delayed until 2015, and the corn grain based ethanol industry is allowed to adjust in response to cellulose-to-ethanol technology introduction.

Results

Under the ETH60 Scenario, the targeted production of ethanol can be achieved for the years 2010, 2020, and 2030. The targeted goals of 1 billion gallons of biodiesel by the year 2012 and 1.6 billion gallons by 2030 can also be achieved. The amounts of ethanol that would be derived from the various feedstocks under the three scenarios are shown in table 1. Under the ETH60 Scenario, through 2012, corn grain continues to be the base of ethanol production. In subsequent years, with the commercial introduction of cellulose-to-ethanol technology, the increase of corn grain for ethanol slows down and remains flat after 2020 at around 14 billion gallons per year. Initially cellulose to ethanol conversion relies on wood residues, but as dedicated energy crops come into commercial production, they become the dominant feedstocks. By 2030, even holding corn grain to ethanol plants at near capacity, less than one in four gallons of ethanol are projected to be derived from corn grain.

Under the ETH60CA Scenario, use of corn reaches a peak in 2012, but with cellulose-to-ethanol technology introduction declines to less than 8 billion gallons by 2030. This suggests excess production capacity in corn grain to ethanol will appear in 2013, and corn grain ethanol plants will likely convert to cellulose or exit the industry. By 2030, the corn grain ethanol industry adjusts, and less than one in six gallons of ethanol are projected to be derived from corn grain.

For the ETH60CACD Scenario, in which commercial introduction of cellulose-to-ethanol technology is delayed, use of corn for ethanol will not peak until 2015 at just under 18 billion gallons. After the peak year, there will be a significant reduction in the use of grain corn, resulting in excess capacity. With a delay in introduction of cellulose-to-ethanol technology, the impacts on the corn grain ethanol industry by 2030 are dampened slightly, about 120 million gallons or about 1.4%, as compared with cellulose-to-ethanol technology introduction in 2012. Also, by 2030, the contribution of corn residues is more significant than under the other two scenarios. Ethanol from corn stover is about 36% higher than the ETH60 Scenario and 12% higher than the ETH60CA Scenario.

In the years beyond 2012, most of the growth in biodiesel production is projected to come from yellow grease and tallow, rather than soybeans. By 2030, 1 billion gallons of biodiesel comes from soybeans, while 0.6 billion gallons is derived from yellow grease and tallow. An alternative target of 2 billion gallons of biodiesel was considered, but to reach this target using soybeans as a feedstock required a price above $8 per bushel.

With a major change in ethanol feedstocks and overall growth in feedstock use, land use patterns would change. For example, under the ETH60 Scenario, dedicated energy crops reach about 34.4 million acres by the year 2030, from very low levels in 2007. Pasture declines from 56.5 million acres to 24.3 million by 2030. Corn acreage increases from 81 million acres and then declines with the introduction of cellulose-to-ethanol technology to around 83 million acres in 2030. About 32.2 million acres of cropland in pasture would return to hay, dedicated energy crops, and other crop production. Acreage planted to soybeans decreases from 73.3 million acres in 2007 to 62.7 million in 2030.

The projected changes in prices of major crops away from baseline levels are shown in table 2. For the ETH60 Scenario, the price estimates indicate that corn, wheat, and soybeans experience a significant price impact. The price impact for corn peaks during the highest period of corn demand for grain ethanol. For the ETH60CA and ETH60CACD Scenarios, the increases in corn prices by 2030 are slightly dampened compared with the ETH60 Scenario, 10 cents per bushel and 2 cents per bushel lower, respectively. With the introduction of cellulose-to-ethanol technology, positive pressure on corn prices is reduced and land is released for production of soybeans. Because the corn grain ethanol industry adjusts under the ETH60CA or ETH60CACD Scenarios, soybean price increases above baseline are lower than under the ETH60 Scenario.

The various sectors within the livestock industry react differently to higher feed prices. Cattle sector impacts are quite different when compared with hog and poultry sector impacts. A reduction in cattle inventories leads to higher prices that offset the sector's increased production costs and reduces the total expenditures on feed. Dried distillers grains (DDG's) can be more heavily incorporated into cattle rations compared with hog or poultry rations.

Under the ETH60 Scenario, there is a projected cumulative increase in net farm income of $210 billion during 2007-30. With these increases in net farm income, decreases in loan deficiency and countercyclical payments are projected. Cumulative reductions in loan deficiency payments and countercyclical payments are projected at nearly $1 billion and $7.8 billion. Hence, the projected cumulative reduction in government payments is $8.7 billion compared with the baseline.

The geographic distribution of cellulosic feedstock production in 2030 for the ETH60 Scenario is presented in figure 1. As shown in figure 1, by 2030, a wide geographic area of the United States contributes cellulosic feedstock. Dedicated energy crops production is concentrated in the Southeast, Southern Plains, and Northern Plains, while corn stover is concentrated in the Midwest.

[FIGURE 1 OMITTED]

Under the ETH60 Scenario, by 2030, a total of $110 billion (2006 dollars) annually is directly generated in the economy via purchasing inputs, adding value to those inputs and supplying biofuels to the nation, with $25 billion from the agricultural sector and $85 billion from the renewable energy sector. About 236,000 jobs are added directly to the agricultural sector, and 58,000 jobs are added directly to the biofuels sector. Including indirect impacts, the estimated economic impacts are $368 billion per year creating an estimated 2.4 million jobs.

Conclusions

The analyses performed indicate that the U.S. agriculture is in a position to play a significant role as a source of energy. For the entire period through 2030, the cumulative displacement could be as high as 10.48 billion barrels of oil, causing a potential reduction in imports of $629 billion. In addition to the ethanol, by 2030, 1.6 billion gallons of biodiesel per year could be produced. Overall, for the period 2007-30, the estimated accumulated gains in net farm income are over $210 billion, and the accumulated potential savings in government payments are estimated to be $150 billion. Due to the geographic decentralization of the production of feedstock, economic gains are projected to accrue in the majority of regions of the country. Significant expansion beyond 60 billion gallons per year would likely require expansion of the region suitable for the production of bioenergy crops, the ability to convert other pastureland (beyond cropland in pasture) into energy crops, allowing the use of Conservation Reserve Program (CRP) acreage for feedstock production, increasing short-rotation wood crops in the Northeast and Northwest regions, increasing yields above those assumed in the analysis, and/or increasing the efficiency of cellulose-to-ethanol conversion. Further research should examine the agricultural, environmental, and economic impacts of changes in one or more these factors.

The Promise and Challenge of Bioenergy (Francis Epplin, Oklahoma State University, Organizer)

References

Aden, A., M. Ruth, K. Ibsen, J. Jechura, K. Neeves, J. Sheehan, B. Wallace, L. Montague, A. Slayton, and J. Lukas. 2002. Lignocellulosic Biomass to Ethanol Design and Economics Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn Stover. Golden, CO: U.S. Department of Energy, National Renewable Energy Laboratory & Harris Group, NREL/TP-510-32438.

BBI International. 2002. "State of Maine Ethanol Pre-Feasibility Study." Report Prepared for Finance Authority of Maine, October.

De La Torre Ugarte, D., and D. Ray. 2000. "Biomass and Bioenergy Applications of the POLYSYS Modeling Framework." Biomass and Bioenergy 18:291-308.

De La Torre Ugarte, D., B. English, K. Jensen, C. Hellwinckel, R. Menard, and B. Wilson. 2006. Dept. of Agr. Econ., The University of Tennessee, December.

English, B., J. Menard, and D. De La Torre Ugarte. 2000. "Using Corn Stover for Ethanol Production: A Look at the Regional Economic Impacts for Selected Midwestern States." Dept. of Agr. Econ., The University of Tennessee.

English, B., J. Menard, B. Wilson, and D. De La Torre Ugarte. 2004. "Integrating IMPLAN with a National Agricultural Policy Model." Proceedings of the 2004 National IMPLAN User's Conference, Sheperdstown, WV, October, pp. 38-47.

English, B., K. Jensen, and J. Menard in cooperation with Frazier, Barnes & Associates, Llc. 2002. "Economic Feasibility of Producing Biodiesel in Tennessee." Dept. of Agr. Econ., The University of Tennessee, December.

Fortenberry, T. 2005. "Biodiesel Feasibility Study: An Evaluation of Biodiesel Feasibility in Wisconsin." University of Wisconsin-Madison, Dept. of Agri. & Appl. Econ. Staff Paper No. 481, March.

Governor's Ethanol Coalition. 2006. "Ethanol From Biomass: How to Get to a Biofuels Future, Recommendations." Washington DC, December.

Harkin, T., R. Lugar, J. Biden, B. Dorgan, and B. Obama. 2007. Biofuels Security Act of 2007. Introduced in Senate 23 IS. January 4.

McAloon, A., F. Taylor, W. Yee, K. Ibsen, and R. Wooley. 2000. "Determining the Cost of Producing Ethanol from Corn Starch and Lignocellulosic Feedstocks." Golden, CO: U.S. Department of Energy, National Renewable Energy Laboratory, and U.S. Department of Agriculture, NREL/TP-580-28893.

McCarl, B., and U. Schneider. 2000. "U.S. Agriculture's Role in a Greenhouse Gas Emission Mitigation World: An Economic Perspective." Review of Agricultural Economics 22:134-59.

McLaughlin, S., D. De La Torre Ugarte, C. Garten, L. Lynd, M. Sanderson, V. Tolbert, M. Walsh, and D. Wolf. 2002. "High-Value Renewable Energy from Prairie Grasses." Environmental Science and Technology 36:2122-29.

Minnesota IMPLAN Group (MIG). 1999. Micro IMPLAN User's Guide. St. Paul, MN, Version 2.0.

Schneider, U.A. 2000. "Agricultural Sector Analysis on Greenhouse Gas Emission Mitigation in the United States." Ph.D. Dissertation, Texas A&M University.

Shapouri, H., J. Duffield, and M. Wang. 2002. "The Energy Balance of Corn Ethanol: An Update, U.S. Department of Agriculture." Washington DC: Office of the Chief Economist, Office of Energy Policy and New Uses (USDA-OCE). Agricultural Economic Report No. 813.

Sheehan, J. 2002. "Life-Cycle Analysis of Ethanol from Corn Stover." Golden, CO: U.S. Department of Energy, National Renewable Energy Laboratory, NICH Report No. PO-510-31792.

Urbanchuk, J. 2001. "An Economic Analysis of Legislation for a Renewable Fuels Requirement for Highway Motor Fuels." AUS Consultants for Governor's Ethanol Coalition.

U.S. Department of Agriculture, Office of the Chief Economist. 2002. "Effects on the Farm Economy of a Renewable Fuels Standard for Motor Vehicle Fuel." Report to Senator Harkin.

Walsh, M., D. De La Torre Ugarte, H. Shapouri, and S. Slinsky. 2003. "The Economic Impacts of Bioenergy Crop Production on U.S. Agriculture." Environmental and Resource Economics 24:313-33.

Daniel G. De La Torte Ugarte is Associate Professor and Burton C. English and Kim Jensen are Professors at the University of Tennessee, Department of Agricultural Economics, 302 Morgan Hall, Knoxville, TN 37996-4518.

This study was partially funded by The Governor's Ethanol Coalition and the National Commission on Energy Policy.

This article was presented in a principal paper session at the AAEA annual meeting (Portland. OR, July 2007). The articles in these sessions are not subjected to the journal's standard refereeing process. Table 1. Ethanol Production from Feedstocks under the ETH60, ETHCA, and ETHCAD Scenarios

2010 2015 2020 2025 2030

Billions of Gallons of Ethanol Corn Grain

ETH60CACD 10.00 15.93 10.78 9.23 8.90

ETH60CA 10.00 9.60 9.15 8.84 8.78

ETH60 10.00 12.96 14.09 14.09 14.09 Wood Residues

ETH60CACD 0.00 1.62 3.77 4.40 5.51

ETH60CA 0.00 4.23 3.75 4.51 5.15

ETH60 0.00 3.63 2.33 4.31 4.54 Wheat Straw

ETH60CACD 0.00 0.00 0.46 0.97 1.77

ETH60CA 0.00 0.55 0.41 1.15 1.70

ETH60 0.00 0.26 0.01 0.97 1.14 Corn Stover

ETH60CACD 0.00 0.00 0.01 5.69 12.10

ETH60CA 0.00 1.82 0.01 8.37 10.76

ETH60 0.00 0.00 0.00 5.91 8.88 Dedicated Energy Crop

ETH60CACD 0.00 0.00 14.40 24.81 32.10

ETH60CA 0.00 3.56 16.66 22.36 34.01

ETH60 0.00 3.40 13.69 19.93 31.71 Total Production

ETH60CACD 10.00 17.56 29.41 45.08 60.37

ETH60CA 10.00 19.77 29.97 45.22 60.39

ETH60 10.00 20.25 30.11 45.20 60.35 Note: ETH60 assumes commercial availability of cellulose-to-ethanol technology by 2012 and the use of corn grain is kept at near production capacity of ethanol plants. ETH60CA allows the corn grain to ethanol industry with cellulose-to-ethanol technology introduction in 2012 (ETH60CA). ETH60CACD delays cellulose-to-ethanol technology until 2015, and the corn grain-based ethanol industry is allowed to adjust. Table 2. Crop Price Impacts Compared with Baseline by Scenario

Changes in Crop Prices from Baseline

Corn Wheat

($/bushel) ($/bushel)

ETH60 ETH60 ETH60 ETH60 Year CACD CA ETH60 CACD CA ETH60 2010 0.86 0.86 0.89 0.11 0.11 0.11 2015 2.05 0.38 0.69 0.46 0.19 0.32 2020 0.19 -0.06 0.36 0.00 0.00 0.07 2025 -0.15 -0.14 -0.04 -0.07 -0.06 0.01 2030 0.59 0.52 0.62 0.36 0.36 0.53

Changes in Crop Prices from Baseline

Soybeans Cotton

($/bushel) ($/pound)

ETH60 ETH60 ETH60 ETH60 Year CACD CA ETH60 CACD CA ETH60 2010 0.75 0.75 0.82 0.00 0.00 0.00 2015 1.42 0.65 1.49 0.02 0.04 0.04 2020 0.75 0.81 0.97 0.03 0.03 0.03 2025 0.54 0.55 0.98 0.03 0.02 0.03 2030 0.89 0.91 1.23 0.02 0.02 0.02 Note: ETH60 assumes commercial availability of cellulose-to-ethanol technology by 2012 and the use of corn grain is kept at near production capacity of ethanol plants. ETH60CA allows the corn grain to ethanol industry with cellulose-to-ethanol technology introduction in 2012 (ETH60CA). ETI I60CACD delays cellulose-to-ethanol technology until 2015, and the corn grain-based ethanol industry is allowed to adjust.