THE EFFICIENT USE OF ELECTRICITY IN BRAZIL: PROGRESS AND OPPORTUNITIES

Howard Geller Gilberto de Martino JannuzziRoberto Schaeffer-Maurcio Tiomno Tolmasquim-May 1997

Planejamento de Sistemas Energticos, Faculdade de Engenharia Mecnica, UNICAMP C.P. 6122, 13083-970 Campinas, SP, Brazil (55-19)239-8420 e-mail: jannuzzi@turing.unicamp.br
--

Programa de Planejamento Energtico, COPPE/UFRJ C.P. 68565, Cidade Universitria, Ilha do Fundo, 21945-970 Rio de Janeiro, RJ, Brazil (55-21)560-8995 e-mail: roberto@ppe.ufrj.br e-mail: tolmasquim@ppe.ufrj.br

American Council for an Energy-Efficient Economy American Council for an Energy-Efficient Economy 1001 Connecticut Avenue, NW, Suite 801, Washington, DC 20036 (202)429-8873

The Efficient Use of Electricity in Brazil, ACEEE

TABLE OF CONTENTS
Abtract.................................................................................................................................... A-1 Introduction ................................................................................................................................ 1 Electricity Conservation and DSM Programs Initiatives............................................................... 3 Residential Sector............................................................................................................ 5 Lighting ............................................................................................................... 7 Water heating....................................................................................................... 7 Refrigeration ...................................................................................................... 10 Other end-uses ................................................................................................... 11 Commercial and Public Services Sectors ........................................................................ 12 Lighting ............................................................................................................. 13 Air Conditioning ................................................................................................ 15 Industrial sector............................................................................................................. 16 Industrial structure ............................................................................................. 16 Motors and motor systems ................................................................................. 18 Overall efficiency potential ................................................................................. 20 Discussion and Recommendations ............................................................................................. 21 Regulations.................................................................................................................... 22 Financing and ESCO Support ........................................................................................ 23 Utility DSM Programs ................................................................................................... 24 Education and Marketing............................................................................................... 25 Acknowledgments..................................................................................................................... 26 References ................................................................................................................................ 26

The Efficient Use of Electricity in Brazil, ACEEE

ABTRACT
This paper reviews the efforts made with electricity conservation and DSM programs in Brazil in the recent past. The principal end uses in the residential, commercial, and industrial sectors are considered. The status of various electricity-saving measures is examined along with initiatives developed by the National Electricity Conservation Program (PROCEL), utilities and other agencies. While some progress has been made, there remains enormous potential for cost-effective efficiency improvements. We conclude with a discussion of strategies for intensifying electricity conservation efforts in the future.

A-1

The Efficient Use of Electricity in Brazil, ACEEE

INTRODUCTION
Electricity use has changed dramatically in Brazil over the past 25 years. Total electricity consumption in Brazil grew from 38 TWh in 1970 to 277 TWh in 1996 (MME, 1988; 1996), an average growth rate of 7.9% per year, or more than twice as fast as that of the corresponding OECD rate of 3.5% per year for 1971-1990 (Pearson, 1996). The share of electricity consumption in the overall energy mix, counting all electricity based on the energy required to generate power in thermal power plants, grew from 16.0% in 1970 to 38.7% in 1995 (MME, 1988; 1996). Figure 1 shows the evolution in electricity demand and electricity demand per unit of GDP for the 1980-96 period. Overall, the electricity intensity of the Brazilian economy increased 63% between 1980 and 1996, or some 3.2% per year on average.

Figure 1: Trends in electricity consumption in Brazil Source: Based on MME (1996). Power sector official estimates traditionally overestimate future electricity demand in Brazil. For example, the decade plan issued by Eletrobras, Brazils national utility holding and coordinating company of the power sector, in 1989 forecast total electricity consumption of 302 TWh in 1995, nearly 15% more than the actual consumption that year. Electricity prices in Brazil fell during the 1970s, 1980s and early 1990s (MME, 1996), encouraging demand growth and creating severe financial problems for electric utilities. However, prices were increased significantly in the mid 90s. The prevailing average electricity tariffs in 1996, excluding taxes, were about US$48/MWh for industrial customers, US$98/MWh for commercial customers, 1

The Efficient Use of Electricity in Brazil, ACEEE

and US$105/MWh for residential customers, or some US$71/MWh for the economy as a whole (Eletrobras, 1997). Thus tariffs are not low by international standards and are adequate to cover operating costs. Even so, they are not adequate to cover the enormous debt service and capital investments required by the Brazilian power sector. To better understand the major electricity conservation opportunities in Brazil, we first need to know how electricity is used in the country. Figure 2 presents estimates of electricity consumption by the major sectors of the economy for the period 1970-1995. The share of the different sectors was relatively constant during this time period. In 1995, for example, the industrial sector accounted for about 48% of total electricity consumption, the commercial sector for 12%, the residential sector for 24%, the public services sector for 9%, and the remaining of the economy for 7% (MME, 1996).

Figure 2: Electricity Demand by Sector in Brazil Sources: Based on (MME 1988; 1996). In the remainder of the paper, we describe the efforts undertaken to stimulate greater efficiency and the results achieved with electricity conservation and DSM programs in Brazil in the recent past, including initiatives developed by the National Electricity Conservation Program (PROCEL), utilities and other agencies. We then look in detail at the status of electricity conservation measures in each sector. Finally, we discuss the main barriers inhibiting energy efficiency improvements and strategies to overcome them.

ELECTRICITY CONSERVATION AND DSM PROGRAMS INITIATIVES

The Efficient Use of Electricity in Brazil, ACEEE

The concern with the efficient use of electricity in Brazil began in the mid-eighties. Led by the power sector, the objective was to reduce the need for new investments due to the power sectors financial problems. The Government of Brazil established PROCEL in December, 1985. PROCEL funds or co-funds conservation projects carried out by state and local utilities, universities, state agencies, private companies, and research institutes. These projects involve research and development, demonstrations, education and training, marketing, direct installation of conservation measures, support of ESCOs, work on legislation, and DSM programs. Also, PROCEL helps utilities obtain low-interest financing for major energy efficiency projects from a low-interest loan fund within the electricity sector. During its initial eight years (1986-93), PROCEL spent a total of about US$24 million on over 100 projects. PROCEL also received an equivalent amount of support for staff, overhead, and travel from Eletrobras. However, Brazil's electric sector experienced severe financial difficulties during the early nineties because of low electricity prices and high debt. Consequently, PROCEL's budget was relatively low and influence relatively limited during 1990-92. A process of renewal was begun in 1993 and continued through 1996. PROCEL's "core budget" of grant funds and staff support increased to around US$10 million per year by 1995-96. In addition, PROCEL arranged about US$21 and US$42 million of financing for utility energy efficiency projects in 1995 and 1996, respectively. PROCEL's major actions are discussed in many of the end use sections below. A comprehensive project review and impact analysis estimated that PROCEL can take credit for about 790 GWh per year of electricity savings due to actions in 1996 alone and about 2,360 GWh per year of electricity savings as of 1996 based on cumulative actions (Geller et al., 1997a). The latter is equivalent to about 0.9% of total annual electricity consumption as of 1996. Considering cumulative actions, about 43% of the savings come from more efficient refrigerator and freezers; 22% from lighting efficiency improvements; 15% from audits, sectoral studies and seminars, and industrial awards; 11% from installation of meters; 7% from motors projects; and 1% from education programs. The electricity savings resulting from PROCEL's activities have been growing very rapidly (see Figure 3). The savings estimate in 1996 is 53% greater than the savings estimate in 1995, considering actions taken annually. Also, savings from actions in 1996 are about four times as large as the savings from actions stimulated by PROCEL as of 1993. The increase in electricity savings is attributed to the rapid growth in PROCEL's budget, projects, and influence during 1993-96, as well as to the cumulative impact of working in some areas for more than a decade.

Figure 3: Electricity Savings Due to the Actions of PROCELa

The Efficient Use of Electricity in Brazil, ACEEE

Does not include PROCEL projects aimed at increasing electricity generation. Cumulative actions refer to savings from projects since PROCEL began in 1986. Source: Based on PROCEL (1997). The 2,360 GWh per year of energy savings produced by PROCEL as of 1996 is equivalent to the power typically supplied by about 565 MW of hydro capacity in Brazil. Assuming an average marginal cost of US$2,000/kW installed (including generation as well as associated T&D investments), PROCEL has reduced supply-side investment requirements by about US$1.1 billion.1 Residential Sector Residential electricity consumption doubled during the seventies, continued to show high growth rates during the eighties, and averaged 6.0% per year during 1990-96. As of 1995, 92% of the households in Brazil were electrified. Growth in residential electricity demand was the primary factor causing total electricity use to rise rapidly in the past three years, and residential electricity demand is expected to continue to rise faster than total demand in the next decade. Currently residential
1

PROCEL also has undertaken some projects to increase power generation at some hydro

plants.

The Efficient Use of Electricity in Brazil, ACEEE

consumption reaches a peak from 6 p.m. to 9 p.m., which contributes significantly to the overall system peak load (see Figure 4).

Figure 4: Load Profile of the South, Southest and Central-west Interconnected Systema
a

Values are average during each hour of 6/26/96, which was one of the days with highest preak demand in 1996. Source: Based on PROCEL (1996). Several residential energy surveys have been conducted since the eighties in various regions of the country (Arouca,1983; CESP, 1986; Freitas, 1996; Graa and Barghini, 1989; Jannuzzi and Pompermayer, 1997; Jannuzzi and Schipper, 1991; and Jannuzzi et. al, 1996). While there are some regional variations in electricity consumption, three end uses are responsible for most of residential electricity use in the country: lighting, refrigeration and water heating. As of early 1997, ordinary residential consumers pay about US$0.15/kWh on average (including taxes). Lower income households are subsidized and receive a discount based on the level of consumption (e.g., 60 per cent discount if consumption is under 50 kWh/month and 40% discount if consumption is 51-150 kWh/month). Efforts have been made to increase the efficiency of electricity use in the three main end-uses mentioned above. Lighting-A national appliance saturation survey conducted in 1988 found a total of about nine lamps per household on average, with 26% of households already using at least one flouorescent lamp

The Efficient Use of Electricity in Brazil, ACEEE

(Geller, 1991). Statistical analysis of this survey estimated that lighting consumes 390 kWh per year per household on average (Lins and Silva, 1996). In spite of the large potential for improvements in residential lighting energy use by substituting fluorescent for incandescent lamps, most of the DSM efforts so far can be classified as small-scale pilot projects. These efforts have been made by several utilities, some with support from PROCEL. Table 1 summarizes the main efforts made during the recent past by three utilities. In spite of the modest size of the residential lighting programs implemented so far, they have contributed to an increase in compact fluorescent lamps (CFL) sales in the country, which increased by a factor of five in the last five years, to about three million lampssold in 1996, which is equal to 1% incandescent lamp sales (Geller et al., 1997b). New larger scale programs are being planned by PROCEL and some utilities to promote the adoption of CFLs in particular regions. In particular, one utility is planning to finance the installation of up to 180,000 CFLs in the residential sector in the city of Fortaleza in Cear state, with consumers paying for the lamps through the monthly utility bill. Another utility is planning a bulk purchase of 135,000 CFLs and giveaway program for low-income households in the city of Vitria da Conquista in Bahia state. In both cases, the local distribution utility is interested in reducing peak demand in order to delay transmission and distribution system investments. Water heating -The most common device used for residential water heating in the country is an electric resistance shower which heats water instantaneously on demand. This device has nominal capacity ranging from 2 kW to 6 kW, and there is a clear trend towards increasing power. The cheapest models available cost US$10-12 and have nominal capacity of 2.0-4.5 kW. Residential water heating is an important end-use in the Southern and Southeastern parts of the country, where more than 85% of residential customers own electric showers. But we also observe a trend for introducing electric water heating systems in warmer areas of the country as well. Electric showers contribute heavily to the overall peak load in the early evening in Brazil. Each electric shower presents a diversified peak demand of 400-500 W on average, but has an average demand during 24 hours of only 20 W; i.e., consuming about 180 kWh per year (Lins and Silva, 1996).

The Efficient Use of Electricity in brazil, ACEEE

Table 1: Main Characteristics of Residential Lighting Programs of Three Different Utilities Utility and year Type of program Lamp costs to customer (US$) Lamp regular prices (US$) Wattage Lamp type Ballast typeb Number of participants Program costs to utility (thousand US$) CEMIG 1990 direct installation none 13 9 and 13 CFL M 514 180 CEMIG 1995/96 direct installation none n.a. 9 CFL M 52,000c 1,100 CPFL 1992 direct installation none 16-22a 22 and 32 CFL and circular E 369 22.2 CPFL 1994 rebate 4-24a 13-34a 15-32 CFL and circular E and M 9,634 550 26,808 CPFL 1995 montly payments 10-25a 13-27a 15-32 CFL and circular E and M n.a. n.a. n.a. CESP 1993 manufacturers discount 11 16 9 CFL M 1,428 19.3 1,350

Total number of lamps 3,000 89,000 (72,000d) 380 Notes: All financial values are in current US$ and refer to the year of the program. a M: electromagnetic, E: electronic. b Refers to the price range of the different models. c Total number of households of the service area invited to join the program. d Number installed as of end of 1996. Sources: Based on Jannuzi (1994); Carvalhes (1996); Jugiwara (1996).

The Efficient Use of Electricity in Brazil, ACEEE

There are a limited range of options that can compete with the electric shower to reduce electricity consumption in domestic water heating systems in Brazil. Urban gas distribution is restricted to few cities in the country and the major use of LPG is for domestic cooking. Use of solar energy for water heating is costly and of limited applicability. The replacement of the electric shower with other technologies is very attractive for utilities, as can be seen from Table 2. Given its high contribution to peak demand, utilities must invest US$800-1,000 per electric shower. If other technologies were used instead, as Table 2 illustrates, the utilities could benefit from avoiding this substantial investment. Table 2: Economics of Replacing Electric Showers for Other Water Heating Systems Costs Water heating systems conventional solar heater Cost of conserved electricity (US$/kWh) Cost of avoided peak capacity (US$/kW) low-cost solar heatera gas-heat on demand electricity cost (peak) conventional solar heater low-cost solar heatera gas-heat on demand Utility 0.02 -0.08 -0.06 0.29 527 -1,815 -902 Consumer 0.54 0.21 0.21 -

Notes: Data refer to 1993 for the State of So Paulo.

a

The system consists of a solar water pre-heater coupled with a low capacity electric shower.

Source: Based on Madureira (1995). For the consumer, on the other hand, in addition to a low initial cost, the electric shower has the advantage of simple plumbing and low installation costs compared to the other alternatives. Given the present tariff level and structure and the prices of the alternative water heating technologies, there is no economic advantage for customers to invest in alternatives to the electric shower. PROCEL, working with one utility (CEMIG, the utility in the state of Minas Gerais), stimulated the development and commercialization of a load control device that prevents the electric resistance heater (and other high power appliances) from operating during the peak period. This device reduces 8

The Efficient Use of Electricity in Brazil, ACEEE

peak demand but does not necessarily save electricity due to consumers altering their bathing schedule. In 1995-96, CEMIG implemented a pilot project where consumers were offered a 20% discount on their electricity rates if they accepted the demand limiters. About 3,000 units were installed in one region of Minas Gerais. The pilot project was a success in terms of both consumer acceptance and cost effectiveness for the utility. Based on this result, PROCEL has encouraged other utilities to start similar projects. In early 1997, seven utilities initiated projects to stimulate the adoption of 40,000 water heater demand limiters. Refrigeration-Refrigerators were used in 75% of Brazilian households in 1995, an increase from 63% saturation in 1988. During 1993-96 alone, nearly 12 million new refrigerators were sold in Brazil, with high sales growth following the adoption of the Plano Real2 in 1994. Most refrigerators sold in Brazil are so-called single door models with a small freezer compartment inside and total volume of 250-340 liters. However, two-door refrigerator-freezers of 300-425 liters are gaining in popularity. Statistical analysis of a national appliance saturation survey conducted in about 10,800 households in 1988 indicated that typical one-door refrigerators in use at that time consumed 525 kWh per year and typical two-door models consumed 803 kWh per year on average (Lins and Silva, 1996). PROCEL along with the appliance manufacturers have made progress in increasing the energy efficiency of new refrigerators in the past decade. PROCEL, working with the manufacturers and the national research institute for the electricity sector (CEPEL), adopted standardized testing and labeling of the electricity consumption of refrigerators and freezers starting in 1986. Due to this initiative and further pressure from PROCEL, manufacturers voluntarily increased refrigerator efficiency by about 10% on average in the late eighties. In the early nineties, PROCEL and the appliance manufacturers negotiated an agreement concerning voluntary efficiency standards which applied to new one-door refrigerators starting in 1995, and to new two-door refrigerator-freezers and freezers starting in 1996. The voluntary standards specify the maximum electricity use as a function of volume, following the approach to standards adopted in the U.S. Approximately half the refrigerator and freezer models being produced as of 1994 (when the agreement was concluded) did not meet the voluntary standards. In 1995, PROCEL began giving a seal of approval and publicity to the top-rated models in the marketplace. In 1996, a pilot project was begun in Manaus, state of Amazonas, involving rebates and intensive advertising to stimulate

The Plano Real was the successful economic stabilization plan which greatly reduced inflation in Brazil.

The Efficient Use of Electricity in Brazil, ACEEE

greater purchase of these top-rated models. Preliminary results indicate that the program has tripled the sales level of the top-rated models in this city. All of these efforts contributed to the gains in efficiency realized during the past decade, as indicated by Figure 5, which shows the range of electricity use for models sold in different years. In 1994, one major manufacturer introduced a new 320 liter refrigerator-freezer which consumed less than half as much electricity as previous models produced and sold in Brazil in this size range. This model immediately won the PROCEL seal of approval. However, approximately 15% of the refrigerators produced in Brazil still contain fiverglass insulation and are relatively inefficient by international standards (Araujo, Cassiolato and Silva, 1997). According to Brazilian appliance manufacturers, new refrigerators as of 1993 consumed 90 kWh per year less on average (about a 15% reduction) compared to new refrigerators produced in 1985. PROCEL estimates this average savings reached nearly 135 kWh per year by 1996, based on the introduction and growing market share for newer high efficiency models (Geller et al.,1997a). Refrigerator and freezer efficiency improvements during 1986-96 resulted in about 2,560 GWh per year of electricity savings as of 1996, about 1% of national electricity use that year. Based on discussions with manufacturers and other experts, PROCEL is taking credit for 40% of this energy savings in its impact analysis (PROCEL, 1997).

Figure 5: Energy Efficiency Improvements in Refrigerators Other end-uses-Use of other appliances such as televisions, room air conditioners, clothes washers, dishwashers, freezers and microwave ovens is increasing in Brazil (Figure 6). For example, about 4 10

The Efficient Use of Electricity in Brazil, ACEEE

million microwave ovens were sold in Brazil during 1990-96, roughly multiplying by seven the saturation of this appliance (Appliance, 1996; ELETROS, 1997). Likewise, the number of stand-alone freezers more than doubled from 2.3 million in 1988 to 5.8 million in 1996. And in fast-growing cities in Northern Brazil (e.g., Manaus or Porto Velho), 30-40% of households already own at least one window air conditioner (Jannuzzi and Pompermayer, 1997). Increasing appliance ownership is contributing to the high rate of growth in residential electricity use observed in recent years. As a result, PROCEL is extending testing, labeling, and promotion of top-rated models to air conditioners, clothes washers and other appliances. Commercial and Public Services Sectors The commercial and public services sectors represent approximately 21% of total electricity consumption in Brazil (Figure 2). Lighting, air-conditioning, and refrigeration are the main end-uses in these sectors, representing about 44%, 20%, and 17% of sectoral electricity use, respectively. There is still a lack of information on the potential for more efficient electricity use in these sectors. Over 600 energy audits were performed in buildings in several regions of the country in the late 1980s and early 1990s, but there has not been a thorough analysis of this data. However, it is clear that the savings potential is large. Furthermore, some progress has been made to increase energy efficiency especially in lighting.

Figure 6: Appliance Sales Trends Source: Based on Appliance (1996); ELETROS (1997).

11

The Efficient Use of Electricity in Brazil, ACEEE

As of early 1997, low-voltage commercial consumers pay US$0.13/kWh on average (including taxes), making electricity conservation very cost effective and of growing interest in the commercial sector. Larger consumers that receive power at 2.3 kV or higher voltages have the option of a timeof-use tariff which includes energy and demand charges that vary depending on the time of day and season. This tariff provides a strong incentive for peak load reduction. Lighting-Some commercial enterprises have increased the energy efficiency of lighting, although there remains tremendous potential for cost-effective efficiency improvements. A number of efficient lighting technologies have been introduced in Brazil in the past decade. Table 3 shows estimates of the number of efficiency measures in use and the resulting electricity savings as of 1996. The total electricity savings from these six measures was about 2,800 GWh per year as of 1996, about 6% of total lighting electricity use or 1.1% of national electricity consumption (Geller et al., 1997b). Most of the savings result from the use of high pressure sodium lamps and CFLs. Table 3: Electricity Savings Resulting from Use of Efficient Lighting Products, 1996 Product CFLs Circular fluorescent lamps T8 lamps Electronic ballasts Specular reflectors High pressure sodium lamps Number in use (million) 7.2 1.2 Average savings (W) 45 60 Average usage (hrs/yr) 2,500 1,000 Savings (GWh/yr) 810 71

5.8 2.0 0.5 2.2

10 28 52 150

3,000 3,000 3,000 4,500 -

173 171 83 1,499 2.807

Total 18.9 Sources: Based on Geller et al.(1997b); PROCEL (1997).

High pressure sodium lamps are steadily growing in acceptance for industrial, street, and outdoor lighting. PROCEL has encouraged the development and use of high pressure sodium lamps in part through its public illumination program. This program has co-funded with local utilities the replacement of inefficient incandescent and self-ballasted street lights with sodium and mercury vapor lamps. As of 1995, about 400,000 lamps had been replaced through this program, resulting in about 12

The Efficient Use of Electricity in Brazil, ACEEE

122 GWh per year of electricity savings at an average cost of saved energy of around US$0.016/kWh (Leonelli, 1996). While most of these substitutions were with mercury vapor lamps, PROCEL has begun to emphasize sodium vapor lamps. Three new projects were approved in 1996-97 for installing about 300,000 sodium lamps, and other large-scale projects are under development. Adoption of CFLs in commercial buildings is growing rapidly in Brazil. CFLs are especially popular in hotels, restaurants, banks, and other types of building that offer a payback of under one year due to the high number of operating hours (Leonelli, 1995). With the recent reduction in import duties, these lamps are now imported, although one multinational company is planning to manufacture CFLs in Brazil. This could significantly reduce the cost of CFLs as a result of lower labor costs, lower taxes and lower transport costs. PROCEL and individual utilities have helped to increase the use of CFLs through co-funding demonstrations, specific incentive programs (see Table 1), energy audits, and general promotional activities. Around 7 million CFLs were in use in Brazil as of 1996, with about 3 million sold that year alone. However, PROCEL and its utility partners are taking credit for stimulating only 10% of the CFLs in use (Geller et al., 1997b). Standard fluorescent lighting is often inefficient and of poor quality, with fixtures that are highly enclosed, lacking reflectors, dirty, etc. The adoption of energy efficiency measures is growing, especially in new buildings. Approximately 6 million T8 lamps, 2 million electronic ballasts, and 0.5 million specular reflectors were used in Brazil as of 1996. Taken together, they saved about 425 GWh per year as of 1996. While these efficiency measures are being disseminated largely through the marketplace, PROCEL and individual utilities have helped to stimulate the introduction and use of these products through R&D and demonstration projects, audits, and other educational and promotional activities. Given the relatively high savings potential and electricity tariffs, adoption of energy-efficient lighting can be very cost-effective for commercial and public sector consumers. Installation of T8 lamps, electronic ballasts, and/or specular reflectors usually have a simple payback period of under two years (Poole and Geller, 1997). In 1994, PROCEL and fluorescent ballast manufacturers signed a voluntary accord calling for a 10% average improvement in the efficiency of electromagnetic ballasts made in Brazil. Tests sponsored by PROCEL showed that this accord had little impact as of 1995. However, the efficiency of major types of electromagnetic ballasts increased about 4% on average in 1996. In 1996, PROCEL and manufacturers agreed to adopt minimum efficiency standards for electromagnetic ballasts. These standards are expected to take effect in 1997 as part of a new mandatory norm. In addition to promoting particular lighting technologies, PROCEL and individual utilities are helping to advance more efficient lighting in commercial buildings through financing projects (e.g, in the Ministry buildings in Brasilia) and promotion of Energy Service Companies (ESCOs). On the order

13

The Efficient Use of Electricity in Brazil, ACEEE

of 30 ESCOs were operating in Brazil as of 1996. These companies were involved in about 125 projects with a total value of roughly US$16 million that year (Poole and Geller, 1997). Air Conditioning-For Brazil as a whole, air conditioning accounts for only about 20% of commercial sector electricity use. But air conditioning can account for more then half of total electricity use in large office buildings, hotels, or shopping centers (Lamberts et al., 1996). A wide range of steps can be taken to reduce electricity use for air conditioning, including: 1) reducing entry of solar radiation and heat gains in buildings, 2) cooling by natural ventilation when the climate permits, 3) increasing air conditioning efficiency, and 4) better control of air conditioner use. Window air conditioners sold in Brazil have been relatively inefficient by international standards due to use of reciprocating compressors and lower cost designs. However, higher efficiency window air conditioners with rotary compressors have been produced in Brazil for export for many years, using imported compressors. PROCEL began testing and labeling the efficiency of window air conditioners in 1995-96, as well as recognizing, awarding, and publicizing the top-rated models. In 1996, one of the leading manufacturers introduced a new line of high efficiency window air conditioners in the domestic market employing rotary compressors. These units have efficiency (EER) ratings of 9.1-10.3 Btu/Wh, 15-30% greater than previous top-rated models. Higher efficiencies can be obtained by using central air conditioning systems in commercial buildings, especially larger systems containing centrifugal compressors (Lamberts et al., 1996). However, it can be more difficult to control air conditioner output in different areas of the building with a central cooling system. Good building design, including limiting the number of north-facing windows, shading windows, using window films, insulation and reflective surfaces, and using natural ventilation, can reduce cooling demand and increase occupant comfort. A study comparing different commercial building designs in various regions of Brazil found that employing a combination of these measures could reduce air conditioning electricity use by 60-75% (Alucci, 1989). In early 1997, PROCEL together with the local utility in Manaus began a air conditioner servicing pilot project. The project involves training of air conditioner contractors and free servicing of package type air conditioning systems used in smaller commercial buildings. The servicing involves cleaning, refrigerant charging, and other low-cost measures. If this pilot project is successful in terms of energy savings and cost effectiveness, utility-sponsored servicing could be adopted on a wider scale. Industrial sector The industrial sector (excluding energy production) accounted for 48% of total electricity use in Brazil in 1995 (Figure 2). Similar to the other sectors of the economy, the overall electricity intensity of the industrial sector rose substantially during the past 26 years. It is estimated that electricity is 14

The Efficient Use of Electricity in Brazil, ACEEE

used within the industrial sector mainly in motors (51%), electrochemical processes (21%), electrothermal processes (20%), refrigeration (6%), and lighting (2%) (Pinhel, 1994). Industrial structure-Growing industrial electricity use has been caused primarily by economic growth and structural shifts, not by declining energy efficiencies (Geller and Zylbersztajn, 1991; Henriques Junior and Schaeffer, 1995). Table 4 presents a breakdown of industrial electricity use by major industry type in 1970 and 1995. As of 1995, the most important industries were nonferrous metals (mainly aluminum) (22% of total industrial electricity consumption), chemicals (12%), iron and steel (11%), food and beverages (10%), paper and pulp (8%), and steel alloys (5%); industries that devote a substantial share of there production for export (Machado, 1996; Tolmasquim et al., 1994). These six industries alone consumed about 68% of total industrial electricity use in Brazil in 1995, up from approximately 60% in 1970. Table 4: Industrial Electricity Use Subsector GWh Iron and steel Steel alloys Nonferrous metals Chemicals Food and beverages Paper and pulp Others 1,927 574 3,414 2,646 1,757 1,667 7,550 1970 % 9.9 2.9 17.5 13.5 9.0 8.5 38.7 100 GWh 14,222 6,406 28,715 14,871 12,727 9,923 40,867 127,731 1995 % 11.1 5.0 22.5 11.6 10.0 7.8 32.0 100

Total 19,535 Source: Based on MME (1988; 1996).

Large industries pay relatively low electricity tariffs, typically US$0.025-0.035 per kWh on average, with the exact price depending on voltage level and load factor. Very large aluminum producers, in particular, have been encouraged to expand smelting facilities in Brazil through special low-cost electricity tariffs. Large industries vigorously resist efforts that would increase energy costs. In spite of this, there has been some improvements in the energy efficiency of these industries, particularly in companies concerned about global competitiveness.

15

The Efficient Use of Electricity in Brazil, ACEEE

Table 5 shows the trends in electricity intensity of the major electricity-consuming sectors since 1975. Two of the largest electricity-consuming sectors--nonferrous metals and chemicals--significantly reduced their electricity intensity during this time period. The increase in specific electricity consumption in some industrial sectors can be attributed to the substitution of electricity for other energy sources rather than to a decrease in efficiency with which electricity is used. For example, between 1975 and 1995, electricity increased its share of total primary energy consumption from 20% to nearly 25% in the iron and steel industry, from over 34% to nearly 43% in the paper and pulp industry, and from 55% to almost 79% in the textile industry, counting all electricity based on the energy required to generate power in thermal power plants (MME, 1988; 1996). Table 5 Specific electricity consumption in selected industrial sectors(MWh/ton) Industrial sectors Cement Iron and steel Steel alloys Chemicals Nonferrous metals Textile 1975 0.119 0.570 5.209 0.622 23.037 1.885 1980 0.120 0.587 5.438 0.481 19.157 2.327 0.824 1985 0.120 0.610 5.653 0.530 17.145 2.905 0.896 1990 0.115 0.627 6.677 0.499 17.677 n.a. 0.899 1995 0.116 0.567 7.182 n.a. 16.060 n.a. 0.878

Paper and 0.798 pulp Source: Based on MME (1996).

Motors and motor systems-Motors account for about half of industrial electricity use, about 40% of electricity use in commercial buildings (mainly through air conditioning, refrigeration, and pumping equipment), and around 40% of residential electricity use (through refrigerators and other appliances). Approximately 1.2 million three-phase and 6.0 million monophase motors are sold annually in Brazil (Soares and Tabosa, 1996). Most of these motors are made by three domestic manufacturers, the largest of which has a 60-70% market share. Brazilian manufacturers started producing high efficiency motors containing silicon steel and other improved features for export in the 1980s (Geller, 1991). The manufacturers began selling these motors in Brazil around 1990. During the early 1990s, motor manufacturers increased the efficiency of both their standard and high efficiency motor lines (Oliveira and Almeida, 1995). Figure 7 compares the efficiency of typical standard and high efficiency induction motors made in Brazil in 16

The Efficient Use of Electricity in Brazil, ACEEE

1996 in the range of 1-200 HP. For smaller size motors, the different in efficiency is 5% or more, while for larger motors the difference is 2-3%. High efficiency motors represented only about 1% of three-phase induction motors sold in Brazil in 1996 (Geller et al., 1997a). Furthermore, high efficiency motors made and sold in Brazil are somewhat less efficient than high efficiency North American motors, as evidenced by comparing Brazilian high efficiency motors with the thresholds used for designating Premium Efficiencymotors in North America (see Figure 7).3

Figure 7: Comparison of the Efficiency of Electric Motors in Brazil and North Americaa
a

Based on three-phase induction motors that operate at 1800 RPMs

Source: Based on Tabosa (1997). High efficiency motors typically cost about 40% more than standard motors sold in Brazil (Soares and Tabosa, 1996). Nonetheless, high efficiency motors are cost effective in many applications when
3

The test procedures for rating the efficiency of motors in Brazil are very similar but not totally identical to the test procedures used in North America.

17

The Efficient Use of Electricity in Brazil, ACEEE

a new motor is needed, with a cost of saved energy of $0.015-0.025 per kWh saved (Soares, Hersztberg and Arouca, 1996). The simple payback period ranges from 1 to 7 years, depending on the size of the motor and the electricity tariff. But the extra first cost, lack of familiarity, and lack of a well-developed delivery infrastructure (i.e., high efficiency motors are not normally stocked by motor distributors) all contribute to the very low market share for high efficiency motors. Besides the low efficiency of electric motors in Brazil, there are two additional factors that add to poor performance of motors in the industrial sector: improper specification (use of oversized motors), and poor operation and maintenance practices. Field surveys in the late 1980s showed that around 71% of all motors in use in industry operate at loads under full capacity, and 25% at loads under 50% of full capacity, resulting in lower efficiencies (Henriques Junior, 1995). The use of energyefficient motors and motor speed controls, along with regular preventive maintenance practices and use of appropriate voltage and power factors, can significantly reduce electricity use in motors. For example, one study estimates a cost-effective savings potential of about 12 TWh (20%) in industrial motor systems after a ten-year DSM effort (Geller, 1991). PROCEL has sponsored a number of projects in the motor systems area, including studies of the level of inefficient practices such as oversizing; demonstrations of adjustable speed drives in applications with a high degree of part load operation; development of higher efficiency motors; adoption of standardized motor efficiency testing and labeling; and development of data bases and software, and guides to help businesses and industries increase motor system efficiency. Efficiency labeling began for 1-200 HP induction motors in 1996. Also, the PROCEL seal of approval was given to the most efficient standard motors sold starting in 1995, leading to some efficiency improvements in new motors in 1996. It is estimated that motor efficiency improvements stimulated by PROCEL in recent years resulted in savings of about 170 GWh per year as of 1996 (Geller et al., 1997a). The savings are limited in part because PROCEL's most influential efforts in this area are relatively recent, and in part because the market share for high efficiency motors is still very low in Brazil. More needs to be done to encourage energy savings in this important end use. Overall efficiency potential-For industry as a whole, PROCELs audit program indicated a savings potential of 8-15% based on low-cost measures like replacing oversized motors, improving transmission systems between motors and subsequent equipment, replacing overloaded lines, adjusting or replacing overloaded transformers, correcting low power factors, correcting electrical problems, and using more efficient lamps or light fixtures (Geller, 1991). Additional savings are possible through the use of energy-efficient motors, variable-speed drives, improving electric furnaces and boilers, electrolytic processes improvements, and cogeneration system, but these measures are more costly (Henriques Junior, 1995). CEMIG, the utility in Minas Gerais state, has promoted more efficient electricity use among their major industrial consumers through information dissemination. CEMIG first performs a detailed audit in one or more companies, followed by seminars for many companies in the same sector in order to 18

The Efficient Use of Electricity in Brazil, ACEEE

disseminate the audit results (Nobre, 1996). CEMIG then collects data on measures implemented and electricity savings achieved by companies that attended these seminars. During 1991-96, CEMIG conducted these studies and seminars in 10 sectors, with a total electricity savings of 161 GWh per year reported (PROCEL, 1997). Some companies have made major strides in increasing their energy efficiency. A large tire production plant owned by Pirelli of Brazil, for example, cut its electricity use per ton of output by 28% between 1992 and 1996 through installing high efficiency equipment, cutting losses in air compressor systems, and other measures including adopting an employee suggestion and cash prize program (Bahia Indstria, 1997). Pirelli, which received a National Energy Conservation Award in 1997, claims this plant in the state of Bahia is now the most energy-efficient of its 74 facilities worldwide.

DISCUSSION AND RECOMMENDATIONS

There is enormous potential for reducing electricity use through cost-effective end-use efficiency improvements in Brazil. Many efficient technologies, such as high efficiency refrigerators, air conditioners, motors, and lighting products are now produced and/or marketed in Brazil. In addition, design and operational changes such as more careful equipment sizing, use of natural lighting and ventilation, and air conditioning, lighting, or water heater control, can reduce electricity use and peak loads. Adoption of these energy efficiency measures and practices is growing, but in most cases has penetrated less than 5% of the eligible market. Energy efficiency improvements are inhibited by a series of market barriers and imperfections including: (a) many decades of economic instability and high inflation, conditions which strongly discouraged life-cycle analysis and longer term investing, and led to purchases based on minimum first cost; (b) due to industrial policy relatively closed markets and lack of competition, which reduce incentives to cut costs; (c) lack of awareness of electricity conservation measures and practices on the part of end users, due in part to the recent introduction of many of these measures; (d) immature energy efficiency delivery infrastructure, again related to the recent introduction and limited adoption of many measures; (e) subsidized electricity prices still paid by large industrial consumers as well as low-income residential consumers;

19

The Efficient Use of Electricity in Brazil, ACEEE

(f) electricity representing a relatively small portion of total costs for most businesses and consumers; (g) lack of capital or attractive financing for many consumers and businesses -- interest rates are generally very high in private markets and borrowing discouraged by heavy bureaucracy, onerous guaranty requirements, etc; and (h) lack of financial incentives for utilities to operate demad-side management (DSM) programs that lead to significant electricity savings by consumers. A number of these barriers have been reduced in recent years. Inflation has greatly decreased and overall economic conditions have improved since the adoption of the Plano Real in 1994. Markets have been opened up and competition is expanding. Many consumers are now paying relatively high electricity prices. And the availability and awareness of efficiency measures is increasing, in part due to the efforts of PROCEL. However, much more could be done to promote efficient electricity use given the savings potential, the high growth in electricity demand in recent years, and the increasing risk of power shortages in Brazil. Indeed, PROCEL has set targets of saving 2.1 TWh per year in 1997 and 2.7 TWh per year in 1998, well above the levels achieved in recent years (see Figure 3). In order to increase the efficiency of electricity use on a wide scale, we recommend that a combination of public policy instruments be used including regulations, financing mechanisms and ESCO support, utility DSM programs, and broad information dissemination and marketing campaigns. Regulations In Brazil, adoption of efficiency standards has been through negotiations of voluntary targets with electrical equipment manufacturers. So far, this approach has given mixed results. In the future, it would be helpful if the government had the authority to adopt mandatory efficiency standards as has been done in North America. Legislation permitting this was introduced in the Brazilian Congress in the early 1990s but has been stalled. This legislation should be updated, reintroduced, and strongly supported by energy planners and policy makers interested in both increasing economic efficiency and reducing the risk of future power shortages. In the absence of legislation specifically setting or calling for minimum efficiency standards, it is possible to adopt minimum efficiency requirements as part of technical norms. These norms are used to certify products and are used by many companies in purchasing goods such as motors or lighting products. But these norms are adopted through a consensus process, and many manufacturers resist stringent efficiency requirements. Good building design is inhibited by the lack of building energy codes in Brazil. Neither the federal government, states or localities have adopted building energy efficiency requirements. The adoption and effective implementation of such codes, including training architects and construction firms and 20

The Efficient Use of Electricity in Brazil, ACEEE

adequately enforcing the codes, is urgently needed. PROCEL has encouraged more energy efficient building design through publishing guides and manuals, funding building audits, demonstrations, and supporting seminars and conferences (Lamberts et al., 1996), but the impact of these activities is uncertain. Financing and ESCO Support The lack of financing and high cost of capital are another barrier that needs to be addressed. One way to do this would be to incorporate energy efficiency requirements as part of financing for other activities like housing projects, industrial expansion, or urban development (Hollanda et al., 1994). In fact, the national development bank (BNDES) is attempting to finance energy efficiency projects in conjunction with loans for industrial expansion (Mello, 1997). As noted above, many ESCOs have begun operating in Brazil. However, there is little experience with performance contracting or third party financing for ESCO projects. Most projects are small, and nearly all projects are being implemented using financing provided by the client. In a few cases, financing of lower cost measures is provided by the ESCO itself (Poole and Geller, 1997). A number of steps are being taken to increase the availability and amount of third party financing for energy efficiency projects. First, PROCEL and local utilities are starting to finance ESCO projects from a low-interest loan fund in the electricity sector. Second, PROCEL is working with the national development banks to establish practical financing for both ESCOs and consumer-sponsored energy efficiency projects. The national development banks have indicated willingness to finance efficiency projects, but financing has been inhibited by relatively small project size, administrative requirements, limited interest on the part of local banks (who process smaller loans for the development banks), and high collateral requirements. Efforts to overcome these barriers should continue. Other actions that could help to expand the ESCO industry in Brazil include establishing an energy efficiency loan guarantee fund, developing and disseminating model energy performance contracts, developing monitoring and savings verification procedures, providing training and independent certification of ESCOs, and publicizing successful ESCO projects. A number of organizations are working on these issues including a Brazilian ESCO association that was started in early 1997. Utility DSM Programs Electric utilities could play a much larger role in promoting more efficient electricity use in Brazil. First, PROCEL should receive additional resources (both funding and staff) in order to stimulate greater adoption of efficiency measures and market transformation in key areas such as lighting, refrigeration, water heating, and motor systems. PROCEL hopes to scale up its activities in 19982001 in part by obtaining a large loan from the World Bank and accompanying grant from the Global Environmental Facility. These funds would be used for state and local implementation programs, pilot projects, and core activities such as training, information dissemination and marketing, and equipment testing and standards. 21

The Efficient Use of Electricity in Brazil, ACEEE

Second, state and local utilities should increase their own end-use energy efficiency programs. A growing number of distribution utilities in Brazil are starting DSM programs. These programs are aimed primarily at reducing peak load in specific regions in order to postpone costly T&D system investments. This strategy has been successfully pioneered by CEMIG in one part of Minas Gerais state (Nobre, 1996). Many distribution utilities in Brazil, however, are still conducting only token end-use efficiency programs due to lack of experiece, inability to recover program costs, and concern about the reduction in sales revenue. While federal regulations allow utilities recover DSM program costs in tariffs, in practice this is not occurring. Furthermore, there is no ability for utilities to recover the net loss revenues or receive a portion of the societal benefits generated by their DSM programs. This situation can and should change in the near future as government-owned utilities are privatized and utility sector restructuring takes place. As part of this restructuring, we recommend that: (a) all distribution utilities be required to set savings targets and operate DSM programs for their customers, working closely with PROCEL; (b) both local DSM programs and PROCEL receive base funding through a distribution system "wires fee" (e.g., 1% of retail revenues), which would serve as a floor but not a cap on utility energy efficiency investments; (c) distribution utilities be allowed to recover costs, net loss revenues, and a portion of societal benefits produced by their DSM programs; and (d) energy savings projects be allowed to participate in energy resource bidding processes conducted by utilities or independent systems operators. Given that three distribution utilities have already been privatized and many more utilities are expected to be privatized in the next few years, it is especially important to provide utilities with financial incentives for operating end-use energy efficiency programs that are in the national interest. Some of these policies have been recommended by the international consultant advising the government of Brazil on utility sector restructuring. Education and Marketing Consumers need to be educated and convinced that increasing energy efficiency is worth the effort, even if energy use represents only a small fraction of the cost of operating a business or household. This barrier exists in all countries and will not be easy to overcome in Brazil, where consumers face a host of problems in addition to those experienced in industrialized nations. On the other hand, if major steps are not taken to both increase electricity supply and improve energy efficiency in the coming years, it is likely that power shortages will occur. This prospect could be used as a "rallying cry" for stimulating consumer action. 22

The Efficient Use of Electricity in Brazil, ACEEE

Raising awareness of energy efficiency opportunities is needed among all types of consumers. Many businesses in Brazil are trying to cut their costs and improve their competitiveness, but are not aware that energy efficiency measures often provide a higher rate of return than other investment options. At the household level, many individuals are used to a culture of waste and fail to take even the most basic actions such as turning off lights and TVs when not in use, in spite of the relatively high electricity prices that now prevail. Likewise, many consumers do not seek efficient appliances or even realize that a range of efficiency levels are available when they are shoppig for major appliances. PROCEL has sponsored educational and information dissemination programs for all types of consumers, and is planning to greatly expand its advertising and marketing efforts starting in 1997. Product labeling and publicity of labels and top-rated products can and should be improved since most appliances displayed in stores do not show energy efficiency labels. Information on successful case studies and best practices could be disseminated to commercial and industrial consumers. Likewise, energy efficiency training is needed for architects, plant and building engineers, and other professionals who influence energy use. And businesses could be encouraged to make high level commitments to increase energy efficiency wherever profitable, following the model of the successful U.S. "Green Lights" program. Finally, it could be helpful to connect efficient electricity use with environmental protection. Concern about the environment has greatly increased in Brazil in the past decade. However, since over 90% of electricity in Brazil is provided by hydropower, there is little recognition that electricity waste is harmful to the environment. But this view is incorrect because hydroelectric dams often have significant adverse environmental impacts (Rosa et al., 1988; Rosa and Schaeffer, 1995), and because new power plants will increasingly depend on fossil fuels. Associating efficient electricity use with a cleaner environment could help to stimulate additional consumer interest and action.

ACKNOWLEDGMENTS
The U.S. Agency for International Development provided the primary funding to support Howard Geller during the preparation of this work. Additional support was provide by the U.S. Environmental Protection Agency, Office of Policy, Planning, and Evaluation and the U.S. Department of Energy. This work was also made possible, in part, by research grants by Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico-CNPq to Gilberto De Martino Jannuzzi, Roberto Schaeffer and Mauricio Tiomno Tolmasquim. The authors appreciate the information and assistance provide by representatives of a number of companies that are producing and/or selling lighting products, motors and appliances in Brazil.

REFERENCES
23

The Efficient Use of Electricity in Brazil, ACEEE

Alucci, M. (1989). Economia de energia eltrica em edificaes comerciais. Paper presented to National Symposium on Energy Conservation in Buildings, So Paulo, Brazil, June. Appliance (1996). Portrait of the Latin American appliance industry. Appliance (January) 85-88. Arajo, J. L. R. H., Cassiolato, J. E. and Silva, D. (1997). Difuso de Tecnologias Eficientes no Brasil. Instituto de Economia, UFRJ, Rio de Janeiro, May. Arouca, M. (1983). Anlise de Demanda de Energia Residencial no Brasil. Masters thesis, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil. Bahia Indstria (1997). Pirelli reduz em 31% o uso de energia, Bahia Indstria, January. Carvalhes, J. B. (1996). Personal communication, Companhia Energtica de Minas Gerais-CEMIG, Belo Horizonte, MG, Brazil. CESP (Centrais Eltricas de So Paulo) (1986). Consumo Residencial de Energia Eltrica e Refrigerao. Report CESP, Rio Claro, SP, Brazil. Eletrobras (Centrais Eltricas Brasileiras) (1997). Boletim Trimestral: Janeiro-Dezembro de 1996. Eletrobras, Rio de Janeiro, RJ, Brazil. ELETROS (1997). Estatsticas ELETROS de Vendas Industriais. ELETROS, So Paulo, SP, Brazil. Freitas, M. V. (1996). Pesquisa de Hbitos de Consumo e de Posse de Eletrodomsticos, e Proposta de Aes para Eficincia Energtica em Santa Maria - DF", Fundao Getlio Vargas, Escola Brasileira de Administracao Pblica, Braslia, DF, Brazil. Geller, H. (1991). Efficient Electricity Use: A Development Strategy for Brazil. American Council for an Energy-Efficient Economy-ACEEE, Washington, DC. Geller, H., Abreu, R. M., Lima, C. E., Lima, M. and Pimentel, G. (1997a). Evaluation of the energy savings due to Brazil's National Electricity Conservation Program (PROCEL). In Proceedings of the 1997 International Energy Program Evaluation Conference, Chicago, IL (forthcoming). Geller, H., Leonelli, P., Abreu, R. M. and Araujo, I. (1997b). Energy-Efficient Lighting in Brazil: Market Evolution, Electricity Savings, and Public Policies. American Council for an Energy-Efficient Economy-ACEEE, Washington, DC, February. Geller, H. and Zylbersztajn, D. (1991). Energy-intensity trends in Brazil, Annual Review of Energy 16. Graa, G. and Barghini, A. (1989). Uso de Eletricidade no Setor Residencial da Cidade de So Paulo. Companhia Energtica de So Paulo-CESP, So Paulo, SP, Brazil. 24

The Efficient Use of Electricity in Brazil, ACEEE

Henriques Junior, M. (1995). Uso de Energia na Indstria Energo-Intensiva Brasileira: Indicadores de Eficincia e Potencial de Economia de Energia. Masters thesis, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil. Henriques Junior, M. and Schaeffer, R. Energy use in the Brazilian industry: gains from energyefficiency improvements. In Proceedings of the ACEEE 1995 Summer Study on Energy Efficiency in Industry: "Partnerships, Productivity and the Environment," Volume II, pp. 99-110, Grand Island, NY, 1-4 August. Hollanda, J. B., Poole, A. D. and Tolmasquim, M. T. (1994). Energy Efficiency and Conservation: Strategic Pathways for Energy Efficiency in Brazil. ESMAP/INEE, Washington, DC. Jannuzzi, G. M. (1994). Brazilian utilities: households in focus. IAEEL Newsletter 3 (6) 6-7. Jannuzzi, G. M. and Pompermayer, M. L. (1997). Uso residencial de eletricidade nas cidades de Manaus-AM, Porto Velho-RO e Macap-AP. Paper presented to XIV SNPTEE-Seminrio Nacional de Produo e Transmisso de Energia Eltrica, Belm, Brazil, 26-30 October. Jannuzzi, G. M. and Schipper, L. (1991). The structure of electricity demand in the Brazilian household sector, Energy Policy 19 (11) 879-891. Jannuzzi, G. M., Leonardi, M. L., Braga, A. and Parente, R. (1996). Pesquisa de Posse de Equipamentos e Hbitos de Uso de Energia na Cidade de Manaus. Convnio FUNCAMP/ Eletrobras no. 73/95, UNICAMP, Campinas, SP, Brazil. Jugiwara, J. K. (1996). Personal communication, Companhia Paulista de Fora e Luz-CPFL, So Paulo, SP, Brazil. Lamberts, R., Lomardo, L. L. B., Aguiar, J. C. and Thome, M. R. V. (1996). Eficincia Energtica em Edificaes: Estado da Arte. Report PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil, March. Leonelli, P. A. (1995). Efficient lighting in Brazil. In Proceedings of the 3rd European Conference on Energy-Efficient Lighting, Newcastle upon Tyne, England, June. Leonelli, P. A. (1996). Personal communication, PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil. Lins, M. P. E. and Silva, A. C. M. (1996). Conditional demand analysis for estimating regional variation in appliance specific electricity consumption for Brazilian household sector. In Proceedings of the 1996 European-Latin American Forum on Energy Research, August. Machado, G. V. (1996). Modificaes Recentes no Padro de Uso de Energia na Indstria Brasileira Face Insero do Pas na Economia Mundial. Masters thesis, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil.

25

The Efficient Use of Electricity in Brazil, ACEEE

Madureira, R. G. (1995). Avaliao Econmica de Tecnologia Solar para Conservao de Energia Eltrica em Aquecimento de gua Residencial. Masters thesis, Faculdade de Engenharia Mecnica, UNICAMP, Campinas, SP, Brazil. Mello, E. B. (1997). Personal communication, BNDES, Rio de Janeiro, RJ, Brazil, May. MME (Ministrio das Minas e Energia) (1988). Balano Energtico Nacional. MME, Braslia, DF, Brazil. MME (Ministrio das Minas e Energia) (1996). Balano Energtico Nacional. MME, Braslia, DF, Brazil. Nobre, E. C. (1996). Experincias da CEMIG em projetos de uso racional de energia. In Proceedings of the VII Congresso Brasileiro de Energia, pp. 2384-2414, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil. Oliveira, A. and Almeida, E. L. F. (1995). Innovation and Energy Conservaton: Electric Motors in Brazil. Unpublished report, Instituto de Economia/UFRJ, Rio de Janeiro, RJ, Brazil. Pearson, P. (1996). Electric power, emissions and economic development. IAEE Newsletter (Winter) 4-7. Pinhel, A. C. (1994). Personal communication, PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil. Poole, A. D. and Geller, H. (1997). The Emerging ESCO Industry in Brazil. American Council for an Energy-Efficient Economy-ACEEE, Washington, DC, March. PROCEL (1996). Plano de Aes do PROCEL/GCOI/GCON - Projeto 1 - Gerenciador da Demanda na Baixa Tenso. PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil, December. PROCEL (1997). Resultados do PROCEL 1996: Economia de Energia e Reduo de Demanda na Ponta. PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil, April. Rosa, L. P., Sigaud, L. and Mielnik, O. (1988). Impactos de Grandes Projetos Hidreltricos e Nucleares: Aspectos Econmicos, Tecnolgicos, Ambientais e Sociais. AIE/COPPE and Editora Marco Zero, Rio de Janeiro, RJ, Brazil. Rosa, L. P. and Schaeffer, R. (1995). Global warming potentials: the case of emissions from dams, Energy Policy 23 (2) 149-158. Soares, J. A. and Tabosa, R. P. (1996). Motores eltricos: uma anlise comparativa de mercado e eficincia. In Anais do VII Congresso Brasileiro de Energia, pp. 2019-2028, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil.

26

The Efficient Use of Electricity in Brazil, ACEEE

Soares, G. A., Herszterg, I. and Arouca, M. C. (1996). Avaliao econmica da utilizao de motores eltricos industriais de induo de alto rendimento. In Anais do VII Congresso Brasileiro de Energia, pp. 2059-2073, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil. Tabosa, R. P. (1997). Personal communication, PROCEL/Eletrobras, Rio de Janeiro, RJ, Brazil. Tolmasquim, M. T., Schaeffer, R. and Geller, H. (1994). Desenvolvimentos recentes da conservao de energia eltrica no Brasil. In O Uso Eficiente da Eletricidade: Uma Estratgia de Desenvolvimento para o Brasil, H. Geller, pp. 195-223, INEE-Instituto Nacional de Eficincia Energtica, Rio de Janeiro, RJ, Brazil.

27