Are biofuels a feasible option?

ARTICLE IN PRESS Energy Policy 37 (2009) 10–14

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

Viewpoint

Are biofuels a feasible option? Jose´ Goldemberg a,, Patricia Guardabassi b a b

˜o Paulo, Avenue Prof. Luciano Gualberto, 1289, Sa ˜o Paulo, 05508-010, Brazil Institute of Eletrotechnics and Energy, University of Sa ˜o Paulo, Avenue Prof. Luciano Gualberto, 1289, Sa ˜o Paulo 05508-010, Brazil Brazilian Reference Centre on Biomass, University of Sa

a r t i c l e in f o

a b s t r a c t

Article history: Received 11 August 2008 Accepted 19 August 2008

Recently a number of objections have been raised against the use of ethanol produced from agricultural products such as maize, sugarcane, wheat or sugar beets as a replacement for gasoline, despite some of their advantages such as being cleaner and to some extent renewable. We address these objections in this paper. Topics discussed include the ‘‘corn connection’’ (which was theorized to be a cause of deforestation in the Amazonia), the rise of food prices due to ethanol production and the real possibilities of ethanol in reducing greenhouse gas emissions. It has been shown that such concerns are grossly exaggerated and that ethanol from sugarcane, as produced in Brazil, is the preferred option for the production of fuel not only in terms of cost but also as a favourable energy balance. Finally, the possibility of expanding ethanol production to other sugar-producing countries is also discussed. & 2008 Elsevier Ltd. All rights reserved.

Keywords: Biofuels Ethanol Sustainability

1. Introduction The present use of ethanol as a fuel is around 3.2 million GJ, accounting for 0.7% of the world’s oil production and 2% of the gasoline consumption, while using less than 1% of the agricultural land in use in the world. In the year 2006, roughly 45 billion litres of ethanol were produced in the world. Three quarters of this was generated in the United States (from maize) and Brazil (from sugarcane), which each country contributing to approximately half the production (Goldemberg, 2007). On technical grounds ethanol is a good alternative to gasoline (Moreira and Goldemberg, 1999). It is produced from agricultural products and does not have the impurities found in petroleum products, such as sulphur oxides and particulates, which are the main source of pollution in large metropolitan areas. In addition, if proper feedstock and agricultural practices are used ethanol reduces greenhouse gas emissions (Goldemberg, 2007). Despite these advantages a number of objections have recently been raised regarding the use of ethanol. Scharlemann and Laurence (2008) argued that, on a complete life-cycle basis, biofuels might have greater aggregate environmental costs than gasoline. Laurence (2007) pointed out a possible ‘‘corn connection’’, linking ethanol production from maize in the United States to Amazonia deforestation. Ziegler (2007), special rapporteur on the Right to Food to the General Assembly of the United Nations 62/2007, raised the issue of the ‘‘potentially grave negative impacts of biofuels (or agro fuels) on

 Corresponding author. Tel.: +55 11 30915054; fax: +55 11 30915056.

E-mail address: [email protected] (J. Goldemberg). 0301-4215/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2008.08.031

the right to food and the serious risk of creating a battle between food and fuel’’. Fargione et al. (2008) and Searchinger et al. (2008), using a worldwide agricultural model to estimate emissions from land use, calculated that, as a result of the expansion of the ethanol production from maize in the United States, some 100,000 km2 of additional land would have to come into cultivation in Brazil, China and India, leading to massive deforestation. We argue here that such concerns are grossly exaggerated and correspond to a very simplistic interpretation of what is really happening in this field.

2. Ethanol from sugarcane and maize To put the problem into perspective one should point out that the land in use for ethanol production 2006 in the United States (from maize) was 51 000 km2 and in Brazil (from sugarcane) 29 000 km2 (Table 1). Together they represent 0.55% of the agricultural area in use in the world, which has over 14 million km2. Despite being small, the expansion of new crops can generate regional problems; in the United States from 2006 to 2007 maize acreage grew by 19% (70 000 km2) to almost 370 000 km2. Most of this expansion came at the cost of soybean planting, which decreased by 17% from 310 000 to 260 000 km2 (50 000 km2) (HGCA, 2008). This is approximately 6% of the world’s area used for that crop, and resulted in prices being driven up (FAO, 2007). For this reason, the point has been made that other countries had increased motivation to expand soybean production, possibly into the Amazonia increasing deforestation (Laurence, 2007).

ARTICLE IN PRESS J. Goldemberg, P. Guardabassi / Energy Policy 37 (2009) 10–14

However, Fig. 1 shows the evolution of area used for grain production, soybean and sugarcane, and indicates that the area used for soybean has not increased since 2004. The reality is that deforestation in the Amazonia has been going on for a long time at a rate of approximately 10 000 km2 per year (INPE, 2008). Therefore, very recent increases are not due to soybean expansion, which has been very small since 2004 (FAO, 2007), but instead due to cattle grazing. It is useful to remember that 930 000 km2 of land are used presently for soy production around the world (FAO, 2007). As a general trend the price of food commodities has been decreasing since 1975, but fluctuations in the area planted and prices of food commodities (as well as crude oil) are frequent, as shown in Fig. 2. Such fluctuations have been taking place for many decades due to an enormous number of factors and events (Naylor et al., 2007). Moreover, not all biofuels have the same impact on food prices; in the case of Brazil, the increased production of ethanol from sugarcane did not lead to an increase in sugar prices. Recent price increases for agricultural products following several decades of declining real prices (von Braun, 2007) are usually seen as one of the causes of famine in the some parts of the world, and give rise to the politically laden controversy of fuel

‘‘versus’’ food, which would affect the poor the most and cause famine in some parts of the world (FAO, 2007). In contrast, the point has been made that higher crops prices will not necessarily harm the poorest people; many of the world’s 800 million undernourished people are farmers or farm labourers, who would ultimately stand to benefit from increased prices (ICTSD, 2008). Regarding cost, the production of biofuels such as ethanol from maize could indeed disturb the price of cereals; however, it is still too early to attribute that cause to recent price variations, and in the process raise new objections to the idea of using biofuels. Table 2 compares the cost of the production of ethanol from maize, wheat, sugar beets and sugarcane in the US, Germany and Brazil. Only ethanol produced from sugarcane in Brazil, which has no subsidies, is competitive as a replacement for gasoline (Henniges and Zeddies, 2004). The reduction in greenhouse gases can be assessed by a lifecycle analysis of the energy balance involved in the preparation of the ethanol. The results are sensitive to assumptions about

2

Harvested area (thousand km ) Area used for ethanol production (thousand km2) Average yield (2003–2006) (metric tons/km2) Total production (2006) (million metric tons) Present production of ethanol (million m3/year) Ethanol yield (m3/km2) World total agricultural arable land

Sugarcane (Brazil)b

c

286 18% or 51d 936c 268c 18.6 365 14 million km2c

c

62 47% or 29c 7400c 455c 17.8e 614

indexed, 2000 = 1

Table 1 Yields and areas of maize and sugarcane for ethanol production (2006) Maize (US)a

11

8

8

7

7

6

6

5

5

4

4

3

3

2

2

wheat

1

1

crude

palm oil soy maize

0 0 1970 1975 1980 1985 1990 1995 2000 2005

a

Naylor et al. (2007). Moreira and Goldemberg (1999). c FAO (2007). d Larson (2007). e Unica (2008). b

Fig. 2. Global trends in process of food commodities and crude oil 1970–2007. Source: Naylor et al. (2007).

600000

500000

Cultivated area (km2)

rice

400000 Grains Soy Sugarcane 300000

200000

100000

0 1990

1992

1994

1996

1998 2000 Year

2002

2004

2006

Fig. 1. Cultivated area in Brazil (1990–2007). Source: CONAB (2006) and IBGE (2006).

2008

ARTICLE IN PRESS 12

J. Goldemberg, P. Guardabassi / Energy Policy 37 (2009) 10–14

Table 2 Comparison of the production costs (h/m3) of ethanol

Total production costa Sale of by products Government subsidies

USA (maize)

Germany (wheat)

Germany (sugar beets)

Brazil (sugarcane)

Rotterdam (gasoline)

394.7 67.1 79.3 248.3

549.7 68.0 – 481.7

595.7 72.0 – 523.7

144.8 – – 144.8

200 – 200

Source: Henniges and Zeddies (2004). a Feedstock represents in all cases 50–70% of total production cost (IEA, 2004).

growing conditions and the use of fertilizers, pesticides and other inputs in the agricultural phase as well as the use of fossil fuels in the industrial phase of production, which involves distillation of ethanol from a dilute solution. In the case of ethanol from maize, the ethanol plants are net importers of processing energy. In contrast, sugarcane distilleries are exporters of energy since most of their energy needs come from the sugarcane bagasse. As a result, the energy balance for maize and for sugarcane is approximately 1.4 and 10.2, respectively (Goldemberg, 2007). Therefore, compared with gasoline, ethanol from maize emits 18% less CO2, while ethanol from sugarcane emits 91% less CO2. Besides the reduction of the emissions of pollutants at all levels including greenhouse gas, the use of other environmental indicators to assure the sustainability of biofuels is desirable, such as avoiding slash and burn deforestation practices as pointed out by Zah et al. (2007) and Searchinger et al. (2008). This assessment is routinely done in sustainable forestry and in the health certification of some products, and should be extended to biofuels without being used as a discriminatory instrument against imports from developing countries.

3. The expansion of ethanol production It is important to point out here that the expansion of sugarcane in Brazil, the second largest producer of ethanol in the world, is taking place over pastureland of which more than 2 million km2 is available, including 100 000 km2 in the State of Sa˜o Paulo that is responsible for two-third of all Brazilian production. Since there are no sugarcane plantations in the Amazonia region leading to deforestation, the only concern is indirect effects such as pushing cattle breeding out of pastureland in the South-eastern states to the Amazonia. This is a real concern but can be avoided by more intensive practices than the ones in use today. The average number of heads of cattle/km2 was 128 in 2001 in the State of Sa˜o Paulo and has increased to 141 in 2005 due to the expansion of sugarcane plantations (Lora et al., 2006). In the country as a whole the density is even lower at closer to 100 cattle/km2. The deforestation in the Amazonia is linked closely with cattle breeding for meat production for internal consumption and export. Today, Brazil has approximately 200 million head of cattle (in 2.37 million km2). The recent US Energy Bill set a target for the production of 56.8 million m3 of ethanol per year by 2015 from maize using ‘‘1st-generation technologies’’, which will most likely require an agricultural area of 140 000 km2. Further expansion of production to 79.5 million m3 is planned, and relies on the use of cellulosic materials involving 2nd-generation technologies, which are still in an experimental phase. The European Union directive will require 14.8 million m3 per year by 2020 to replace 10% of the gasoline consumed, but it currently produces only 1.55 million m3 per year (mainly from sugar beets).

Brazil uses 85% of its production domestically, while 15% is exported to the United States. Brazilian ethanol production will likely double by 2012/2013—36 million m3 per year, replacing approximately 50% of the gasoline that otherwise would be used in the country. If the use of cellulosic materials does not materialize in large scale until 2022, ethanol consumption mainly from ‘‘1st-generation technologies’’ will triple to 113.6 million m3 per year, not considering the 79.5 billion litres expected from cellulosic materials. Production of ethanol from maize and ethanol on the basis of 1st-generation technologies will be at least 108 million m3 per year, up from 34 million in 2006, as shown in Table 3. Expanding ethanol production from maize in the United States will face severe obstacles, as 18% of the present maize acreage of 370 000 km2 has already being dedicated to ethanol industry and affecting soybean production. Furthermore, the production from cellulosic materials, which could be a solution, is still facing technological problems likely to remain unsolved until 2015. However, increases in productivity (including genetic modification) might help significantly by reducing the additional land needed. In all likelihood, though, most of the ethanol required would have to be imported from countries in the Southern Hemisphere, such as Brazil, where available land and climate are particularly favourable for ethanol production from sugarcane (Mathews, 2007), and production can take place without any further deforestation, as recognized by Searchinger et al. (2008). There are almost 100 countries producing sugarcane covering an area of 200 000 km2 (approximately 0.5% of the total world area used for agriculture). Table 4 lists the 15 most important producers representing 86% of the total production of sugarcane. It is easy to convert plants producing sugar to ethanol distilleries, and most of the 325 existing plants in Brazil have a dual purpose. Only 47% of the sugarcane area of this country (29 000 km2) is used to produce ethanol. It is clear therefore that the production of ethanol from sugarcane could be expanded significantly if the example of Brazil is followed by several other countries that use a fraction of their sugarcane for ethanol. Colombia already has four large distilleries in operation, and other sugar-producing countries, mainly in the Caribbean, have plans for more (Nogueira, 2007).

4. Conclusion A worldwide increase of 50% of the present area dedicated to sugarcane (100 000 km2) by 2022 (up from the 29 000 km2 presently in use in Brazil) would result in the production of 79.5 million m3 of ethanol, which together with the United States production would more than suffice to meet projected needs. Carbon emissions will be reduced by approximately 57 million tons per year (Goldemberg, 2007).

ARTICLE IN PRESS J. Goldemberg, P. Guardabassi / Energy Policy 37 (2009) 10–14

13

Table 3 Ethanol demand projections 1st-Generation technologya Maize (US)b

2006 2012/2013 2015 2020h

Sugarcane (Brazil)c

European Uniond

Area used (thousand km2)

Cubic meters (million)

Area (thousand km2)

Cubic meters (million)

Cubic meters (million)e

51f – 130 –

18.6 – 56.8 –

29g 49 – –

17.8 36 – –

1.55 – – 15

2nd-Generation technology

2022

US

Brazil

European Union

79.5 litres (billion)

–?

–?

a China, India and Australia adopted E5 or E10 mandates in a number of provinces as well as Canada, Argentina, Philippines and South Africa by 2010–2012. One estimates that by 2022 not less than 7.6 million m3 per year will be needed in these countries (REN21, 2006). b Productivity in 2006: 365 m3/km2 maize. c Productivity in 2006: 614 m3/km2 sugarcane. d Biofuels Research Advisory Council (2006). e 2004 data. f 18% of total maize area in the US; 3.2% of total maize area in the world. g 13% of total sugarcane area in the world; 47% of the sugarcane area in Brazil. h Present EU production of 2 billion litres/year, mainly from sugar beets.

Table 4 Main sugarcane producer countries (2006) Country

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Brazila India China Thailand Pakistan Mexico Colombia South Africa Australia Cuba Philippines Indonesia United States of America Vietnam Argentina Total

Area harvested (thousand km2) 61,529.29 42,000.00 12,200.00 9,362.27 9,073.00 6,682.93 4,257.34 4,200.00 4,150.00 3,971.00 3,922.80 3,700.00 3,634.50 2,851.00 2,846.39

Sugar production (thousand tons) 31,450 30,140 12,855 6720 3615 5633 2445 2300 4822 1150 2232 1900 3438 2440

174,380.52

Source: FAO (2007). a Only 47% of the sugarcane area is used to produce ethanol.

Worldwide, the arable area in use has grown 0.24% per year in the last 44 years, or approximately 30 000 km2 per year, and this trend is expected to continue (FAO, 2007). Presently, there are high import duties on ethanol imports to protect local industries in Europe and the United States, which will eventually be removed to allow the European Union and the United States to reach their demand projections, thus contributing to a reduction of greenhouse gas emissions from gasoline. The expansion of the use of ethanol from sugarcane as a replacement to gasoline has clear environmental advantages not only in reducing CO2 emissions but in improving the quality of the air in large metropolitan areas such as Beijing, Mexico City and Sa˜o Paulo, where automobiles are the main source of pollution. Importing ethanol from the many countries that could produce it, particularly in the Caribbean area, by the United States and

European Union could be a mutually beneficial solution, reducing costs to consumers in these countries and generating hundreds of thousands of jobs in the producing ones.

Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at doi:10.1016/j.enpol.2008.08.031

References Biofuels Research Advisory Council, 2006. Biofuels in the European Union— A vision for 2030 and Beyond, 40pp. (29.7  21.0 cm), ISBN:92-79-01748-9. ISSN:1018-5593. Available at /http://ec.europa.eu/research/energy/pdf/biofuels_vision_2030_en.pdfS. CONAB (National Supplying Company), 2006. Agricultural Database, available at /http://www.conab.gov.br/conabweb/download/safra/BrasilProdutoSerieHist.xlsS. FAO (United Nations Food and Agricultural Organization), 2007. FAOSTAT, available at /http://faostat.fao.org/default.aspxS. Fargione, J., Hill, J., Tilman, D., Polasky, S., Hawthorne, P., 2008. Land clearing and the biofuel carbon debt. Science 319, 1235–1238. Goldemberg, J., 2007. Ethanol for a sustainable energy future. Science 315, 808. Henniges, O., Zeddies, J., 2004. Competitiveness of Brazilian ethanol in the EU. HGCA, 2008. North American crop update, 2007. Available at /www.openi.co.uk/ h070724.htmS posted by 24 July 2007 and consulted on 11 February 2008. IBGE (Brazilian Geography and Statistics Institute), 2006. Municipal agriculture production, temporary and permanent crops. Available at /ftp.ibge.gov.br/ Producao_Agricola/Producao_Agricola_Municipal_[anual]/2006S. ICTSD (International Centre for Trade and Sustainable Development), 2008. Biofuels production, trade and sustainable development: policy discussion. Draft Paper. IEA (International Energy Agency), 2004. Biofuels for transport: an international perspective. INPE (National Institute for Spatial Research), 2008. Prodes Project—satellite monitoring of Amazon Forest. Annual estimative 1988–2007. Available at /http://www.obt.inpe.br/prodes/prodes_1988_2007.htmS. Larson, E., 2007. Prospects for second generation biofuels technologies. In: Conference on biofuels: an action for a less carbon-intensive economy. 4–5 December 2007, EPE-UNCTAD, Rio de Janeiro, Brazil. Available at /http:// www.unctad.org/Templates/Meeting.asp?intItemID=1942&lang=1&m=14692&year=2007&month=12S. Laurence, W.F., 2007. Switch to corn promotes Amazon deforestation. Science 318, 1721.

ARTICLE IN PRESS 14

J. Goldemberg, P. Guardabassi / Energy Policy 37 (2009) 10–14

Lora, B.A., Monteiro, M.B., Assunc-a˜o, V., Frigerio, R., 2006. Levantamento Georreferenciado da Expansa˜o da Cultura de Cana-de-Ac- u´car no Estado de Sa˜o Paulo (Georeferenced Assessment of Sugarcane Culture Expansion in Sa˜o Paulo state) Project Report, unpublished, Sa˜o Paulo (in Portuguese). Mathews, J.A., 2007. Biofuels: what a biopact between North and South could achieve. Energy Policy 35, 3550–3570. Moreira, J.R., Goldemberg, J., 1999. The alcohol program. Energy Policy 27, 229–245. Naylor, R.L., Lisha, A.J., Burke, M.B., Falcon, W.P., Gaskell, J.C., Rozelle, S.D., Cassman, K.G., 2007. The ripple effect—biofuels, food security and the environment. Environment 49, 30–43. Nogueira, L.A.H., 2007. Biocombustı´veis na Ame´rica Latina—a situac- a˜o atual e perspectivas. REN21, 2006. Renewables global status report 2006—update. Scharlemann, J.P.W., Laurence, W.F., 2008. How green are biofuels? Science 139, 43–44.

Searchinger, T., Heimlich, R., Houghton, R.H., Dong, F., Elobeid, A., Fabiosa, J., Togkoz, S., Hayes, D., Yu, T., 2008. Use of US croplands for biofuels increases greenhouse gases through emissions from land use change. Science 319, 1238–1240. UNICA, 2008. Ethanol production in Brazil. Time series 1990–2007. Available at /www.portalunica.com.br/portalunica/files/referencia_estatisticas_producaobrasil-9-Tabela.xlsS. Von Braun, J., 2007. When foods makes fuel: the promises and challenges of biofuels. Keynote Address. Biofuels, Energy and Agriculture. Zah, R., Bo¨ni, H., Gauch, M., Hischier, R., Lehmann, M., Wa¨ger, P., 2007. Okobilanz von energie production: okologische bewertung von biotreibstoffen (Ecological Balance of Energy Production: Ecological Evaluation of Biofuels). Swiss Confederation, Bern, Switzerland. Ziegler, J., 2007. Draft report of the special rapporteur on the Right to Food to the General Assembly, 62/2007T.