Brazil first used ethanol as a transport fuel in 1903, and now has the world's largest bioethanol programme. Since the creation of the National Alcohol Programme (ProAlcool) in 1975, Brazil has produced over 90 billion litres of ethanol from sugarcane. The installed capacity in 1988 was over 16 billion litres distributed over 661 projects (Anon, 1989). In 1989, over 12 billion litres of ethanol replaced about 200,000 barrels of imported oil a day and almost 5 million automobiles now run on pure bioethanol and a further 9 million run on a 20 to 22 per cent blend of alcohol and gasoline (Hall et al, 1992a) (the production of cars powered by pure gasoline was stopped in 1979). From 1976 to 1987 the total investment in ProAlcool reached $6,970,000 million and the total savings equivalent in imported gasoline was $12,480,000 million (Anon, 1989; Lima, 1989).

Table 5 shows sugarcane, sugar, and alcohol production as well as sales of alcohol-powered cars in Brazil from 1976/77 to 1990. Sugarcane production has increased from about 88 million tons in the 1976/77 harvest to 241 Mt in 1988/89, while alcohol production has increased from about 660 million litres to 12.3 billion litres during the same period (Mazzone, 1989). The planted area of sugarcane increased from 1,540,000 ha in 1972 to 2,820,000 ha in 1982 and 4.31 million ha in 1987 (IBGE, 1989). In the 1983/84 harvest, 47 per cent of the total amount of sugarcane crushed was used for ethanol production, and in 1987/88 the proportion reached 61 per cent (Rosillo-Calle et al, 1991). In 1978, only 0.3 per cent of vehicles sold in the domestic market ran on ethanol but by 1985 this figure has reached % per cent However, by 1990 this was down to an estimated 50 per cent due to difficulties in alcohol supply and retail prices (Mazzone, 1989).

Apart from ProAlcool's main objective of reducing oil imports, other broad objectives of the programme were to protect the sugarcane plantation industry, to increase the utilization of domestic renewable-energy resources, to develop the alcohol capital goods-sector and process technology for the production and utilization of industrial alcohols, and to achieve greater socio-economic and regional equality through the expansion of cultivable lands for alcohol production and the generation of employment Although ProAlcool was planned centrally, alcohol is produced entirely by the private sector in a decentralized manner.

Table 5. Sugarcane, sugar, and alcohol

 

Sources: Compiled from Mazzone, 1989

Notes:

^a The years correspond to a twelve-month period from 1979 through 1990.
^b Percentage of alcohol-powered cars manufactured in a year.
^c Estimated

Achievement of these objectives involved an intricate and politically difficult combination of technology and economic policy in the agricultural and industrial sectors. The Government established several support mechanisms: agreeing that the State-owned oil company, Petrobras, would purchase a guaranteed amount of ethanol; providing financial incentives such as low interest rates and $US2.0 billion in loans to ethanol producers; establishing a price policy to ensure an effective remuneration to alcohol producers; encouraging consumers by selling ethanol at 59 per cent of the price of gasoline (the price of ethanol is now 75 per cent of the price of gasoline, and hence there is almost no subsidy in ethanol (Goldemberg et al, 1992)); public investment in agricultural research; and incentives for alcohol production and utilization technology, and for private innovation and investment. In the Centre-South region of Brazil, it is no longer considered necessary to provide economic incentives to producers and end-users. The combination of these factors, together with the introduction of new equipment, increases in productivity and the number of new plantations, has given a continued boost to alcohol production.

1. Improved sugarcane yields

Between 1977 and 1985, sugarcane yields per hectare increased by 16 per cent and yield of alcohol per ton of cane increased by 23 per cent (Goldemberg et al, 1992) with alcohol productivity increasing from 2660 litres/ha to 3800 litres/ha (Mazzone, 1989). Cane productivity stood at 62.6 tons/ha in 1988 (Carpentieri, 1991). New varieties of sugarcane have been developed in order to gain greater productivity on poor soils, in consideration of the need to expand sugarcane growth without interfering with traditional food production. Intercropping and rotation cropping technologies developed by Brazilian sugar research laboratories have also made it possible for sugar plantations to increase food production. The Brazilian Government has undertaken a major effort to promote the improvement of cane-production technology, e.g., through the establishment of cane prices based on sucrose content rather than on weight, efficient management of sugarcane plantations and pest control, among other devices.

2. Industrial development

The ProAlcool programme has accelerated the pace of technological development and reduced costs within agriculture and other industries. Brazil has developed a modem and efficient agribusiness capable of competing with any of its counterparts abroad. The alcohol industry is now among Brazil's largest industrial sectors, and Brazilian firms export alcohol technology to many countries. Brazilian distillery manufacturers have been so successful in producing efficient hardware that the distilleries were frequently found to perform at up to 30 per cent above their rated capacity. In a number of distilleries, small additional investments have increased production still further to 50 per cent above nominal capacity (Weiss, 1990).

Another industry which has expanded greatly due to the creation of ProAlcool is the ethanol chemistry sector. Installed capacity for ethanol utilization in the chemical industry rose from 60,105 tons in 1976 to 336,980 tons in 1984. From 1975 to 1985 the ethanol-based chemical sector consumed a total of nearly 2.2 billion litres of ethanol, i.e., about 3.5 per cent of the annual alcohol production (Rosillo-Calle, 1986). Ethanol-based chemical plants are more suitable for many developing countries than petrochemical plants because they are smaller in scale, require less investment, can be set up in agricultural areas, and use raw materials which can be produced locally.

3. Social development

Until recently the Government's main objectives had been economic growth, with relatively little emphasis on social development. However, rural job creation has been credited as a major benefit of ProAlcool because alcohol production in Brazil is highly labour-intensive. Some 700,000 direct jobs with perhaps three to four times this number of indirect jobs have been created, although it is unclear how many of these jobs are new ones. The investment to generate one job in the ethanol industry varies between $12,000 and $22,000, about 20 times less than in the chemical industry for example (Goldemberg et al, 1992).

4. Environmental impacts

Environmental pollution by the ProAlcool programme has been a cause of serious concern, particularly in the early days. The environmental impact of alcohol production can be considerable because large amounts of stillage are produced and often escape into waterways. For each litre of ethanol produced the distilleries produce 10 to 14 litres of effluent with high biochemical oxygen demand (BOD) stillage. In the later stages of the programme serious efforts were made to overcome these environmental problems, and today a number of alternative technological solutions are available or are being developed, e.g., decreasing effluent volume using the Biostil process and turning stillage into fertilizer, animal feed, biogas etc. These have sharply reduced the level of pollution and in Sao Paulo. The use of stillage as a fertilizer in sugarcane fields has increased productivity by 20-30 per cent (Rosillo-Calle, 1986). Tougher environmental regulations have also reduced pollution considerably.

5. Economics

Despite many studies carried out on nearly all aspects of the programme, there is still considerable disagreement with regard to the economics of ethanol production in Brazil. This is because the production cost of ethanol and its economic value to the consumer and to the country depend on many tangible and intangible factors making the costs very site-specific and variable even from day to day. For example, production costs depend on the location, design and management of the installation, and on whether the facility is an autonomous distillery in a cane plantation dedicated to alcohol production, or a distillery annexed to a plantation primarily engaged in production of sugar for export. The economic value of ethanol produced, on the other hand, depends primarily on the world prices of crude oil and sugar, and also on whether the ethanol is used in anhydrous form for blending with gasoline, or used in hydrous form in 100 per cent alcohol-powered cars.

The costs of ethanol were declining at an annual rate of 4 per cent between 1979 and 1988 due to major efforts to improve the productivity and economics of sugarcane agriculture and ethanol production (Goldemberg et al, 1992). The costs of ethanol production could be further reduced if sugarcane residues, mainly bagasse, were to be fully utilized as has been shown by Ogden et al, (1990). With sale credits from the residues, it would be possible to produce hydrous ethanol at a net cost of less than $0.15/litre, making it competitive with gasoline even at the low early-1990 oil prices. Using the biomass gasifier/intercooled steam-injected gas turbine (BIG/STIG) systems for electricity generation from bagasse, they calculated that simultaneously with producing cost- competitive ethanol, the electricity cost would be less than $0.045/kWh. If the milling season is shortened to 133 days to make greater use of the barbojo (tops and leaves) the economics become even more favourable. Such developments could have significant implications for the overall economics of ethanol production.

Several economic analyses exist for Brazilian alcohol production. Zabel (1990) estimated that the average cost of ethanol was $0.20 litre in Sao Paulo State and that this could be reduced by 17 per cent by the year 2000 to $0.16 litre. Reddy and Goldemberg (1990) calculated that the cost of ethanol produced in Sao Paulo was about $0.185 per litre; and that at this price, ethanol could compete successfully with imported oil. Goldemberg et al (1992) estimate that an overall cost reduction of about 23 per cent could be achieved in a few years just by using existing technology. Borges and Campos (1990) calculated that ProAlcool is economically feasible with both basic and high petroleum prices but in the low-petroleum-price scenario their analysis indicates that it would be more economical to use gasoline instead of ethanol.

The total gains per annum of ProAlcool were estimated to be equivalent to $7.2 billion and $2.5 billion in the high and basic petroleum price scenarios, while in the low-petroleum-price scenario there would be an estimated loss of about $700 million.

In 1989, ProAlcool received severe economic criticism, with many voices calling for its reduction or even total dismantling (Seroa-da-Motta, 1989). The main reason was the sharp drop in oil prices and me steady growth in Brazil's domestic petroleum production, e.g., about 212 million barrels in 1988 at an estimated cost of $20 per barrel versus an “estimated cost” of alcohol of $45 per barrel according to Young (1989). With the current relatively low oil prices (about $19-20 per barrel, May, 1993) and increased domestic oil production, the economics of alcohol production are unfavourable on a microeconomic basis.

The situation was exacerbated due to a shortage of alcohol fuel. In 1988, the Government was forced to import 200 M litres of alcohol to fill the fuel-ethanol shortage which continued in 1989 and 1990, resulting in a drastic reduction in sales of alcohol cars. Although a drought in the north-east was blamed for the shortfall, part of the crisis stemmed from bureaucratic and technocratic shortcomings. For example, fuel consumption was greatly stimulated when prices were frozen by the Summer Plan (a Federal economic anti-inflation programme launched in January 1989) and by the stagnation of alcohol prices, both of which discouraged the distillers from producing more alcohol. Government policies of maintaining low alcohol prices relative to the cost of gasoline were not accompanied by sufficient incentives to expand the production of alcohol, while depressing the price to hold down inflation destabilized the industry further. This led to the Government raising prices in 1991. Further, in 1989, sugar prices surged in (he international market making it more profitable to sell sugar than to produce alcohol. Meanwhile, farmers had been shifting from cane to other more profitable export commodities so sugar production was down. This illustrates the vulnerability of the programme to short-term market fluctuations (Hall et al, 1992a; Goldemberg et al, 1992).

Many of the problems with ProAlcool are blamed by the industry on two institutions: the Institute of Sugar and Alcohol (IAA), which in June 1989 finally lost its 56-year monopoly over the country's sugar industry; and Petrobras, the State oil monopoly. The first was blamed for creating a price production stalemate by not correctly evaluating price levels for sugarcane and alcohol, and the second, for its long-standing opposition to the alcohol programme, e.g., payment delay tactics to distillers which, in an economy with very high inflation rates, results in large financial losses to the alcohol producers (Mazzone, 1989). Petrobras' opposition to ProAlcool stemmed from the fear of losing its monopoly of liquid-fuel supply and also because the company made large investments in fluid catalytic crackers (large refining installations that convert residual fuel oil into lighter distillates, especially gasoline) which increased the proportion of gasoline obtained from oil and resulted in large surpluses of gasoline.

6. Future implications

The recent economic problems have decreased the overriding importance of ethanol as a liquid-fuel substitute although supply interruptions and energy security are still of great concern to Brazil. Some policy changes were contemplated including a reduction in the alcohol content in gasoline from 22 per cent to 18 per cent, a decrease in the retail differential between gasoline and alcohol from 40 per cent to 25 per cent, and a proposed reduction in the manufacture of alcohol-powered cars to between 30 to 50 per cent of all new vehicles. These last two measures posed the greatest long-term threat to the future of ProAlcool. It is likely that the trend to lower use of ethanol-fuelled cars will continue, considering present low oil prices and the Government's attempt to reduce subsidies for ethanol production.

Brazil's alcohol programme crisis has many international as well as domestic implications. Abandoning ProAlcool, or even down-playing it, could mean a great increase in Brazil's capacity to produce sugar which could have serious implications for the world sugar trade. It was ProAlcool in part, that pushed international sugar prices upward in the 1980s after a decade of slump. A new export policy would flood the international sugar market, which would have serious economic consequences for the cane-growing developing countries which are already calculated to be losing $7 billion in sugar export earnings annually as a result of trade barriers in industrialized countries (Durning and Brough, 1991).

However, ProAlcool is an outstanding technical success that has achieved many of its aims, its physical targets were achieved on time and its costs were below initial estimates. It has enabled the sugar and alcohol industries to develop their own technological expertise along with greatly increased capacity (Rosillo-Calle, 1990). It has increased energy independence, made significant foreign-exchange savings, provided the basis for technological developments in both production and end-use, and created jobs. Overall, Brazil's success with implementing large-scale ethanol production and utilization has been due to a combination of factors which include: government support and clear policy for ethanol production; economic and financial incentives; direct involvement of the private sector, technological capability of the ethanol production sector, long historical experience with production and use of ethanol; cooperation between Government, sugarcane producers and the automobile industry; an adequate labour force; a plentiful, low-priced sugarcane crop with a suitable climate and abundant agricultural land; and a well established and developed sugarcane industry which resulted in low investment costs in setting up new distilleries. In the specific case of ethanol-fuelled vehicles, the following factors were influential: government incentives (e.g., lower taxes and cheaper credit); security of supply and nationalistic motivation; and consistent price policy which favoured the alcohol-powered car (Hall et al, 1991a; Goldemberg et al, 1992).

The Brazilian experience with ProAlcool shows the inherent difficulty of long-term energy planning. As Weiss (1990) points out

 

“on the positive side, Brazilian energy planners enjoy a substantial buffer against any possible future energy shortage, an advantage the rest of the world may some day come to envy. But as things now stand, the ProAlcool Program appears today to be an expensive and impossible-to-cancel insurance policy against an unlikely contingency”.