Printer Friendly

A return to the past? Brazil's privatization of public utilities: the case of the electric power sector.


In the first half of the 1990s Brazil's privatization program concentrated on directly productive state enterprises, and by May of 1997 the government had sold 53 firms for a sum of US$ 26.125 billion. Most of these firms were located in the steel, petrochemical, fertilizer and mining sectors.(1) In the second half of the decade the focus of privatization was on public utilities - electric power generation, transmission and distribution; telecommunications, railways, highways, ports, etc. In 1996 Light (the power company supplying Rio de Janeiro), the Via Dutra (the main highway linking Sao Paulo and Rio de Janeiro), the Rio de Janeiro-Niteroi bridge, and some railways lines were privatized through long-term concessions, and plans for the privatization of most other public utilities were under way for the rest of the decade.

With the shift towards public utilities, a larger participation of foreign capital in the country's privatization process was expected. While external investments played an important role in the privatization experience of most other Latin American countries, they played a minor role in Brazil. By early 1996 their share of total privatization sales amounted to 16.3%. This was expected to change over the rest of the 1990s with the increased emphasis on public utility privatization.

To what extent will the privatization of public utilities lead to conditions prevailing prior to the 1950s, when they were in the hands of private firms, many of them of foreign ownership? What were the reasons for the large early presence of foreign capital in these sectors and what were the reasons for their nationalization? What are the reasons for a return to the private sector? Will this return revive some of the problems such private firms had in the previous period, or have changing economic and political circumstances eliminated earlier sources of conflict?

To deal with these questions we shall concentrate on the electric power sector and divide this paper into the following sections. We shall first summarize some of the conceptual problems a country faces when dealing with public utilities. This will be followed by a review of the pre-1950 experience with private, especially foreign, capital in this sector and the circumstances that led to its nationalization. There follows a discussion of the rise and fall of public ownership of the sector, which will lead to a review of the plans in the second half of the 1990s to privatize this public utility. We shall conclude by considering the challenges which a privately-run public utility sector faces in Brazil's post-import substitution era.


Public utilities (electric power generation, transmission and distribution; telephone; roads; ports, etc.) have traditionally been viewed as natural monopolies to be controlled either by the state or to be regulated when in the hands of private firms. In the latter case, the ownership of the utility can belong to a private firm whose output is regulated by a government agency or commission? Another model of a private utility involves no private property, but rather a concession granted by the state for a finite period of time (30 or 40 years) during which a private firm performs a specific service. According to Guislain and Kerf, with a concession "... the state ... delegates to the private sector the right to provide a service, yet retains some control over the sector by incorporating in a concession contract or license the terms and conditions - including the rights and obligations of the service provider - that will govern the infrastructure project..."(3) A concessionary contract includes references to rates to be charged and the method by which periodic readjustments can take place, and may also include provisions for investments to modernize and expand the specific services involved.(4)

The major concern of a private firm is to be able to charge a tariff which will enable it to cover not only current operational costs, but also generate a surplus to expand and technologically improve its operations, and to have a rate of return which will satisfy its owners. Another important concern is that the concession's duration is long enough for a firm to recoup its investments. The interest of the state is to see the infrastructure services grow in consonance with the rest of the economy (avoiding shortages which raise production costs) and to provide those services at what it (and the public) consider to be reasonable prices. The aim of a regulating agency is to reconcile the concerns of the parties.(5)

When infrastructure services are provided by state enterprises, one can envision a situation where they are independent of the government's central policymaking apparatus, which leads to a situation similar to one described above with privately run services. If, however, these government firms are subservient to the central state apparatus, the policymakers will have to decide whether to allow a pricing pattern which produces resources to cover the firm's current and capital expenditures, or whether low infrastructure prices will prevail in order to subsidize various consumer and industrial groups. These subsidies can either by financed by taxing certain sectors of society or by contributing to the government's budget deficit which, if financed by the central bank, will result in inflationary pressures.

As we shall see, the Brazilian experience over the last 100 years has covered all these scenarios. Our emphasis will be on the power generation and distribution sector.


In the 19th century Britain was the major investor in Brazil, accounting for 78% of total foreign investments.(6) This share declined in the first three decades of the 20th century, but even by 1930 Britain still accounted for half of foreign capital located in Brazil. Most of Britain's investments went into railroads and public utilities. The latter two accounted for 70% and 12% of total investment respectively in 1885; for 58% and 16% in 1905; 44% and 41% in 1913. In the decades after World War I the United States overtook Britain as the major investor, and by 1951 it accounted for 44% of total foreign investments, declining again to 30% in 1991; Britain's share declined to 12% and 7% in those years; while Canada's share was 30% in 1951, falling to 6% in 1991;(7) Germany's share, which was negligible prior to World War II, accounted for 14% in 1991.(8)

In 1929 the share of public utilities in total foreign investments still amounted to 50%, falling to 47% by 1940. After World War II the nationalization process caused this share to fall to 15% by 1952 and 2.4% by 1992. Manufacturing accounted for 24% of foreign investments in 1929, rising to about 75% in 1992.

To attract foreign capital in the 19th century the Brazilian government made use of legislation which guaranteed a minimum rate of return, ranging from 6 to 7%, for a period of 60 years. Railways and sugar processing plants were the sector that most benefited from these incentives. The rate of return guarantee "... meant that a major difficulty faced by foreign capital invested in an economy outside the gold standard was partly circumvented, as the actual rate of return did not vary with fluctuations in the exchange rate."(9) Toward the end of the century these guarantees became a political and economic liability to the government, resulting in the gradual nationalization of railroads.


In the years 1880 to 1900 electricity was introduced in Brazil by small private firms which provided power for the illumination of public places and for various types of economic activities (mining, small manufacturing enterprises, and the handling of some agricultural products), and most of the energy was thermal in origin.(10) Until the 1930s all concessions were granted by municipal governments. Foreign capital made its appearance at the turn of the century. In 1897 two business associates, who were politically well connected, had obtained a 40 year concession from the municipality of Sao Paulo to provide electric tram services. They made contacts in Canada to obtain financing and technical help. Their potential North American partners urged them to negotiate a broader concession, and this led to an agreement which included additional tram lines and also their entrance into the field of power generation and distribution. The year 1899 saw the establishment of the Toronto based Sao Paulo Railway, Light and Power Company Ltd., a decree in Brazil authorizing that company to function in Brazil, and the transfer of the concessions of the two businessmen to the new Canadian company.(11)

Soon after Light entered the Sao Paulo transportation and electric power market, it overwhelmed smaller local competitors by either absorbing or by eliminating them. In 1904 the Canadian company founded the Rio de Janeiro Tramway, Light and Power Company. In 1905 it acquired a concession, which had been previously granted to a smaller firm, to furnish electricity through the use of hydroelectric power. Soon Light monopolized Rio de Janeiro's power, gas, tramway and telephone system. Some local firms in Rio de Janeiro tried to oppose Light's entrance into the market, but Light used its influence with the U.S. State Department to persuade the Brazilian authorities to go through with the concession. This would not be the end of Light's conflict with local groups. In 1909 Light's Sao Paulo concession was challenged by a local group (Companhia Docas de Santos, which controlled the Santos Harbor facilities), but Light had enough political influence with Sao Paulo's municipal council to maintain its position.(12)

Although Light dominated the electric power supply in the cities of Sao Paulo and Rio de Janeiro, other cities were supplied by small locally-owned firms and by 1920 343 electric power firms were operating in Brazil on a concessionary basis with municipalities. Their owners were usually local merchants or landowners who obtained local concessions.(13)

The 1920s were characterized by the construction of larger power plants and a process of concentration of concessionary firms, especially foreign-owned ones. Light bought out concessions in many municipalities of the states of Sao Paulo and Rio de Janeiro. Also participating in the consolidation of the sector was American and Foreign Power Co. (Amforp), a subsidiary of Electric Bond and Share. It penetrated the interior of Sao Paulo state and other parts of the country, buying controlling shares of many local companies, and gradually introducing uniform voltage and modernizing transmission lines. By 1930 the power supply of almost all of the most developed regions of the country was dominated by Light and Amforp.(14)


Until 1930 Brazil's federal government intervened very little in the sector. In 1904 it issued a decree (no. 5.407) establishing rules for concessionary contracts for firms using hydroelectric sources. It contained such principles as: concessions would not be exclusive; maximum concessionary period would be 90 years; and after the termination of the concession, the granting organ would get control over assets without indemnification; and the revision of tariffs would occur every 5 years. The general impact of the decree was mild. Light's concession allowed it to revise its tariff more frequently as its concession called for the application of the so-called "gold clause," which permitted the firm to automatically readjust tariffs when the exchange rate was devalued. The resulting fluctuations in tariffs caused frequent protests from many of the affected sectors.

Until 1930 governmental power in the electric energy industry resided to a large extent with municipal governments.


Beginning 1931 the federal government began to make drastic institutional changes. All previous arrangements were suspended, including the old "gold clause." Both the precarious conditions of the Brazilian economy as the great world depression spread and also the impact of a nationalistic campaign (led by the industrialist Eduardo Guinie, whose group had the concession to operate the port of Santos, but who had lost in the electric power sector to Light)(15) against the behavior and profits of foreign investors in the public utilities sector led to a decree of July 1934, known as the "Codigo das Aguas" ("Water Code"), which became the basic legal instrument for the federal government to regulate the water and electric energy sectors. Private groups in the latter could now operate only under a concession granted by the federal government for a duration of 30 years.(16) At the end of this period the assets of the concessionaire would revert to the state (with or without compensation). The Codigo was the first comprehensive legislation on power in Brazil. It set up a mechanism to supervise firms receiving a concession, ensuring adequate service at "fair" tariffs. The latter would be fixed, taking into account costs, depreciation, and a reasonable rate of return on investments. The latter was based on historical cost of capital, and allowing a rate of return of 10%. This would be the main bone of contention in the following years, as Light and Amforp would constantly ask for a rate based on replacement cost.(17)

Although the principle of tariffs being based on historical cost was part of the law, it did not play a role in determining tariffs, due to both political pressures and to bureaucratic difficulties in implementing it. However, during the rest of the Vargas regime, including his "Estado Novo" dictatorship from 1937 to 1945, no new concessions were given to foreign companies. Government regulation of the sector was entrusted to a National Council for Water and Electric Energy until 1960, when this was transferred to a newly created Ministry of Mines and Energy. During World War II tariffs were frozen and only in 1945 did the government permit an increase of 10% in order to finance a wage increase of workers in the sector. (18)

With the decline of investments in the sector during the 1930s and 1940s, Brazil began to experience increasingly severe power shortages, which led to long periods of power rationing. Given the increasing uncertainty over future returns on investments, both Light and Amforp had little incentive to finance major projects.(19) The Cook mission to Brazil in the first half of the 1940s, which was supposed to help plan the mobilization of resources for the war effort, pointed to the electric power sector as being one of the major bottlenecks for the country's industrial growth and recommended greater planning and the interconnection of various power systems. These recommendations resulted in 1946 in a National Electrification Plan, which emphasized region interconnections. The Plan also called for investments to be concentrated in small and medium-sized power plants, with the state acting as coordinator.


The growth of the public sector as the generator and supplier of electric energy started on a gradual basis. In the mid-1940s the state of Rio Grande do Sul transferred municipal concessions to the state. The same occurred in the state of Sao Paulo. In the northeast the federal government founded CHESF (Companhia Hidreletrica de Sao Francisco), which was to construct a power generating plant at Paulo Afonso, and transmission lines to supply power to the country's northeast.

In the 1950s, with the emphasis on Import Substitution Industrialization (ISI) as the principal growth strategy, the government adopted the position that the huge investments in energy generation and transmission required to accompany the urban-industrial growth of the country would take place through government enterprises (both federal and state), while distribution was to be left to the private sector. In 1953 a Federal Electrification Fund was proposed to finance investments in the sector. Its resources were to come from a tax on energy consumption and from the newly created National Bank for Economic Development (BNDE). In 1962 the coordination of the growing state electric sector was transferred to the new holding company, Eletrobras.(20)

In the mid-1950s an official study group still saw an important role for private foreign companies in the expansion of the power generating capacity of the country if there were an improvement in the concessions, such as monetary correction of the historical investment cost, a rise in the allowed remuneration from 10% to 12%, a more flexible system of tariff adjustments, etc. Nationalist reactions to this proposal were so strong as to make such adjustments not viable.(21) The net result was that the expansion of power generation was left entirely to the public sector (both federal and state).(22)

The growth of the state in electricity generation manifested itself in the construction of the hydroelectric works at Paulo Afonso, owned by CHESF, which supplied power to the northeast region and came into operation in 1955. Its capacity was increased in the second half of the 1950s. In the Center-South, the federal government created the Centrais Eletricas de Furnas in 1957. Although Light and Amforp were initial shareholders, their participation declined in successive increases in capital and by the early 1960s their share had fallen to 2%. Construction began in 1958 and the enterprise came on stream in 1963.

Another project of the federal government was CHEVAP (Cia. Hidreletrica do Vale do Paraiba), founded in 1960 to provide electricity to the city of Rio de Janeiro and came into operation in 1967. Most notable was CEMIG (Centrais Eletricas de Minas Gerais) of the state of Minas Gerais, whose most important investment was the hydroelectric dam of Tres Marias on the Sao Francisco river, coming on stream in 1962. In Sao Paulo, the incapacity (or unwillingness) of Light and Amforp to attend to the growing demand for power in the region led the state government to directly enter the sector. In 1953 the state created USELPA (Usinas Eletricas de Paranapanema), construction began with a loan from the World Bank, and in 1958 the first turbines went into operation. Its power also served the state of Parana. In 1961 the government created CELUSA (Centrals Eletricas the Urubupunga) and its generators were ready to supply the city of Sao Paulo in 1969. In 1966 all Sao Paulo state companies were united into CESP (Centrals Eletricas de Sao Paulo).(23) The southern state of Rio Grande do Sul took over the subsidiaries of Light and Amforp without compensation. Many other states also established state-owned electric power companies to handle the growing demand.(24)
Table 1a. Brazil: Installed Electric Power Capacity (M-W)

1925                      507
1930                      779
1935                      850
1940                    1,244
1945                    1,342
1952                    1,985
1955                    3,148
1960                    4,800
1965                    7,411
1970                   11,233
1980                   31,147
1996                   55,130

Source: IBGE, Estalisticas Historicas do Brasil (Rio de Janeiro:
IBGE, 1990); and IBGE, Anuario Estatistico do Brasil 1994.


Although Light and Amforp increased their generation capacity, they did so at rates substantially below the increase of demand. One thus observes a notable decline of their share of installed capacity.(25) It will be noted in Table 1 that the private sector's share of installed power capacity declined from 82.4% in 1952 to 55.2% in 1962 and 33.6% in 1965.

The nationalization of Rio Grande do Sul in 1959 had substantial repercussions both within and outside of Brazil. Political pressures were growing to expand this nationalization and there was an increased inclination by foreign groups to sell out. In April 1963 Amforp made an agreement with the Brazilian government to sell its installations for USS 135 million. Of this sum, 75% would be reinvested in other sectors and 25% could be remitted abroad in U.S. dollars. Before this agreement was consummated, however, there occurred the military coup of 1964. The new government made an agreement to buy out Amforp for the same sum. The latter would be lent to Eletrobras at a yearly interest rate of 6.5%. The Light group was bought out in 1979.

From the 1970s on the sector had, for all practical purposes, been taken over by the state and in the subsequent decades gigantic investments were made to expand generation capacity, including the building of the world's largest hydroelectric project - the Itaipu dam, built in collaboration with Paraguay's state electric company - the Tucurui dam in the north, and the planning of nuclear power plants through an agreement with German firms in 1975. Substantial emphasis was also given to the interconnection of various state-owned companies and the building of transmission lines.


Brazil's installed electric power capacity during the period of state dominance of the sector expanded from 6,355 MW in 1963 to 42,860 MW in 1984 and 52,741 MW in 1993. It is interesting to note that despite the increasingly sophisticated planning method of Eletrobras, most investments in the 1970s were concentrated on power generation, while resources going into transmission and distribution were limited. This created serious sectoral and regional disequilibria in the 1980s.(26)

In the first two years of its existence Eletrobras was financed through a tax on electricity consumption and an occasional compulsory loan. At first, the latter amounted to 15% of the value of the sale to customers, subsequently rising to 20%. Additional finance was obtained through a 10-year bond issue, paying 12% interest. The tax on energy consumption was lower for rural activities, residential consumers and industries.

By the early 1990s the sector was coordinated by Eletrobras. The system had four regional utilities and about 60 state and local companies. Eletrobras was the "holding company" of four regional companies which were engaged in generation and transmission and accounted for about 57% of total generating capacity; it also controlled some state-level utilities, operating mainly in the distribution area: Light of Rio de Janeiro and Escelsa in Espirito Santo. The remaining generating capacity was in the hands of companies owned by state governments, in which Eletrobras had minority participation, and there were a sprinkling of small companies owned by municipalities and private investors, which operated mainly in the distribution field.(27)


Given Brazil's inflationary problems, it is not surprising that the government gradually came to use its enterprises as tools of macroeconomic policy to try to control the general rise of prices. This was especially the case of the electricity sector. Allowing tariff increases below the rise in the general price level contributed not only to a decline of internally generated investment funds, but often produced losses which forced the government to provide subsidies. This, in turn, increased government expenditures and worsened government budget deficits which were usually financed in an inflationary manner.

In the first years of the military governments electric tariffs were increased to make up for past price distortions. From 1964 to 1969 electricity tariffs increased by 62.4% per year, while the price level increased by 39% per year, and in subsequent years (until 1973) tariffs continued to be readjusted slightly ahead of inflation. This policy enabled the sector to increase internally generated resources from 34% in 1967 to 44.9% in 1973 (see Table 2).(28)

From the late 1960s on, the now state-dominated electric sector relied increasingly on loans from foreign banks. This was related to the rising costs of investments in large power generating dams (such as Itaipu) which were far away from the consuming centers, and thus also called for large investments in transmission facilities. Even if real tariff had not declined, they could not have generated the necessary finance for such large projects.
Table 2. Brazil: Origin of Resources of the Electrical Sector
(% distribution)

                             1967     1973     1979     1984

Internal                     34.0     44.9     24.2     17.9
Forced Loans                  8.1      9.4      7.6      3.9
State Resources              31.9     20.3      6.1      6.0
Domestic Loans               13.0      6.6     30.1      9.4
Foreign Financing            13.0     18.8     32.0     62.8

Total                       100.0    100.0    100.0    100.0

Source: Panorama (1988), pp. 218 and 225; Politicas (1995), p. 169.

In the period following the world oil price shock of 1973, which was an important contributory factor to the rising rate of inflation in subsequent years, the government forced electric companies to follow tariff policies which were linked to its macroeconomic objectives rather than the needs of the sector.(29) In 1974 the federal government ordered an equalization electric tariff throughout the country, which led to a transfer of resources from profitable to loss-making state enterprises.(30) Also, tariff regulation became increasingly complex due to a wide range of discounts and special tariffs " attain objectives in income distribution, energy policy, load and demand management and industrial and economic policies."(31) From 1975 on the government limited tariff adjustments to rates inflation, and thus, as can be seen in Table 3, real tariffs declined steadily for the rest of the decade. Table 2 shows how, as a result of these policies, internal generation of funds declined steadily over the next decade, while reliance on foreign finance rose dramatically. It should be noted that the latter not only reflected the results of the use of electric tariffs as a policy tool to control inflation, but it also reflected the attempt of the Brazilian government to obtain foreign exchange by forcing its enterprises to borrow abroad sums which they did not need.(32) The sector's foreign indebtedness worsened dramatically when international interest rates more than doubled in 1979-80 and its debt servicing in real terms rose by 117% from 1978 to 1990.(33)

During the 1980s the capacity of Brazil's public sector to invest in infrastructure deteriorated progressively. In the power sector yearly investments declined from US$ 8.3 billion in 1982 to US$ 3.2 billion in 1993.(34)


The continuing fiscal crisis of the Brazilian state in the mid-1990s led to the decision by the government to expand the privatization process to public utilities. It had been estimated in 1995 that public infrastructure investment needs for the following five years amounted to US$ 71.7 billion, which would imply yearly public investment expenditures of between US$ 14.3 and 17.9 billion. As in 1995 only expenditures of US$ 7.3 billion had been contemplated, it became obvious that only the private sector could raise such resources.(35)
Table 3. Brazil: Average Real Electricity Tariff (1964 = 100)

1963               91       1978              112
1964              100       1979              106
1965              140       1980              106
1966              144       1981              120
1967              156       1982              114
1968              138       1983              101
1969              150       1984               98
1970              166       1985               93
1971              163       1986               97
1972              175       1987              102
1973              164       1988              101
1974              145       1989               71
1975              150       1990               73
1976              129       1991               72
1977              118

Source: Politicas (1995), p. 146; Historia e Energia (1996), p. 69.

The privatization of public utilities followed the tradition prevailing prior to the nationalization process, i.e., the private sector (domestic and foreign) was allowed to operate public utilities on the basis of concessions. The constitution of 1988 provided the basis for concessions instituted in the 1990s and the Concessions Law (Lei das Concessoes) of 1995 regulated article 175 of the Constitution, establishing the rules under which the state could delegate public services to the private sector.(36) The Concessions Law states that the winning firm of a concession has the obligation not only to render adequate services, but to upgrade the technology as demand expands. It also contains four important provisions:

1. That all concessions are granted for a fixed period of time, at the end of which there will be an open competition for a new concession;

2. That there will be no government subsidies;

3. That the user partake in supervising services; and

4. That in granting new concessions there will be no guarantee of a fixed remuneration based on total costs.

However, the law also states that tariff revisions will consider the evolution of a concessionaire firm's costs.(37) For the electrical sector the Chamber of Deputies created a new regulatory agency in 1996 - the National Electric Energy Agency (Aneel).

Two other important changes were introduced in the 1990s. In 1993 Sintrel (National System for the Transmission of Electric Energy) was established. This new institution makes it possible for any private firm which has energy creating capacity to have access to the national transmission grid. This will increase competition among power generators in serving the public throughout the country. Laws were also passed in 1995 which eliminated the differentiation between private domestic and foreign firms, and thus made it possible for the latter to bid for concessions in the electricity sector.(38)

The aim of the government was to create competition among private firms in the generation of electricity and in its distribution, while the transmission lines would remain in the public sector.

At the beginning of the 1990s Brazil had 62 firms operating in the electrical energy sector. Both federal and state-owned firms dominated the sector. Power generation was dominated by the federal government with 59% of capacity, 36% of capacity in the hands of state-owned firms, 4.7% representing self-generating capacity of private firms, and 0.3% consisting of private firms and municipal governments.(39) There were 23 privately-owned firms which were located in the distribution sector and only accounted for 2% of total sales. Almost all firms were interconnected to a national grid whose extension was 1.5 million kilometers.

The first privatization of the electricity sector occurred on July 11, 1995 with the sale of 51% of the shares of the Eletrobras subsidiary Escelsa (Espirito Santo Centrais Eletricas) for US$ 399.9 million at the Rio de Janeiro stock exchange. This was a price at 11.8% above the minimum price which had been set for the sale. The buyers were banks (led by Pactual, Icatu, Opportunity, Bozano Simonsen, Nacional, and GTD) and 11 pension funds of state companies, who paid 66% in cash and the balance in 'junk bonds' issues by the government.(40) Escelsa is a public utility company involved in power generation, transmission and distribution, spanning 90% of the state of Espirito Santo. In December 1994 it had 2,602 employees. It is Brazil's second largest electricity distributor, and at the time of its privatization charged the second lowest tariff in the country.(41) The winning consortium did not include a foreign group and had little experience in the power sector.

Light was privatized the following year on May 21, 1996. Slightly over 54% of its shares were sold at an auction on the Rio de Janeiro stock exchange for a total of US$ 2.3 billion. It was the largest sale in the country since the beginning of Brazil's privatization program in 1990. About 70% was paid in cash and the rest in outstanding government bonds. The buyers were a combination of foreign and domestic firms, with the federal government participating through its development bank (BNDES), obtaining 9.14% of total shares.(42) The major foreign buyers were the French state firm Electricite de France (EDF), and the U.S. firms, Houston Power Industries and American Corporation (AES), each obtaining 11.4% of total shares. Other major buyers were the recently privatized Cia. Siderurgica Nacional (7.25%) and GTD (1.41%) which represented various pension funds. In addition, 10% of the shares still owned by Eletrobras were reserved for Light's employees, who could purchase these shares through BNDES financing. The shareholders agreed to let EDF manage the company.

The fact that foreign investors were willing to pay cash encouraged the federal and state governments of Brazil to plan for the sale of other utility assets. This was also the first time in the privatization program that non-Brazilian investors had submitted the winning bid and won operating control of a large utility company.

CERJ (Companhia de Eletricidade do Rio de Janeiro) was privatized through an auction on November 20, 1996. This was the first time that a company owned by a state government was sold to the private sector. The winning consortium was led by the Chilean company Chilectra. Also included in the consortium were Endesa of Spain and EDP of Portugal. CERJ distributes power to 57 towns, has 4.2 million customers, and 4,382 employees. Chilectra paid US$ 605 million, which was 30% over the minimum price which had been set by the state government. Besides the cash price, the Chilectra group committed itself to invest US$ 500 million over a two-year period.(43) The terms of the auction gave the new owners annual inflation-based tariff increases for a period of 8 years, and profitability would depend on the capacity of the new owners to streamline the bloated bureaucracy into an efficient organization.(44)

The fourth electric company to be privatized was the Companhia Eletrica da Bahia (Coelba). It was auctioned off at the end of July 1997 for a sum of R$ 1.73 billion (about US$ 1.57 billion), which was about 77.3% above the minimum stated price. The winning consortium was led by a Spanish company Iberdrola, in association with some Brazilian retirement investment funds.

With the sale of the four companies the proportion of the electricity market in private hands rose from 2% in January 1995 to close to 20% in August 1997. In 1997-98 the government planned to sell 11 subsidiaries of Eletrobras and the state government of Sao Paulo had plans to privatize Eletropaulo, which was responsible for 22% of Brazil's and 69% of Sao Paulo's electricity demand.


As the privatization of the electrical sector proceeds and concessions are regulated by Aneel (Agencia Nacional de Energia Eletrica), which replaced DNAEE in 1996, will Brazil witness the development of a new institutional arrangement which will result in the continued harmonious growth of the sector, i.e., a growth that accompanies the demand for electricity by the country's industries and cities, and that is adequately balanced between generation, transmission and distribution?

The privatization of Brazil's public utilities in the second half of the 1990s was taking place within a context of an economy which had undergone a liberalization process since the beginning of the decade. Given the constant pressure for a fiscal adjustment and the worldwide movement towards market orientation and the retreat of the state in economic affairs, and Brazil's need for foreign capital to boost the country's investment ratio, public utility privatizations took place in a "foreign investor friendly" atmosphere. This was especially obvious as the privatization movement in the electric power industry was gaining momentum. For instance, Eletrobras reported in November 1996 that its profits in the first nine months of 1996 were 1,201% greater than in the same period of 1995. The president of Eletrobras was reported to have stated that one explanation for this performance was the increase of tariffs above the rate of inflation.(45) This performance certainly set a favorable stage for the privatization of other entities of Eletrobras. Also, two days after its privatization, it was announced that CERJ would be allowed to raise its tariffs by 10%. It was stated in the concession contract that CERJ would be able to have yearly tariff adjustments based on the general price index.

Other measures which contributed to the private investor-friendly atmosphere was the abolition in 1993 of the Uniform National Tariff system; Ordinance 337 of April 1994 which created a unified transmission grid which was to lead to the development of an open access system for all users; the above discussed concession law of 1995; the IPP (Independent Power Producers) Law of 1995, which allows independent producers to sell power to third parties and thus created competition with public-sector utilities.

In the old regulatory environment electricity prices were uniform on a nationwide basis and the government guaranteed a return on assets of 10-12% per year. Those companies (of course, all public) making more than 12% had to transfer excess profits to those unable to make at least 10%. Under the new regulatory law all companies are responsible for setting their own rates, based on their cost structure, which then has to be approved by the regulatory agency Aneel.(46)

The stage was thus set for the evolution of a competitive electricity market. New consumers whose load exceeds 3 MW can choose their suppliers, while current users with a minimum of 10 MW load, and are connected to the transmission grid, can buy energy from independent power producers. Within three years the latter will be free to choose from any non-federal generating company, while within five years consumers with a minimum of 3 MW load and connected to the grid will be allowed to buy electricity from any utility, IPP or self-generator.

However, some aspects of the old regulatory regime continue. Ownership of power plants continues to revert to the government when the concession expires; all sales of energy to utilities must be done by a federal generating company; the purchase of Itaipu's energy is still compulsory for the utilities.

The basic problem for Brazil's public utilities in general, and the power sector in particular, was that privatization was taking place without sharply defined regulatory framework. It has been claimed that the reason foreign investors did not submit bids in the privatization of Escelsa was due to the lack of clarity, transparency and predictability of the tariff setting mechanism. For the privatization of Light the National Privatization Council tried to modify and clarify the regulatory process in order to attract foreign capital. New rules were designed to give more control to the principal buyer by eliminating regulators' day-to-day management responsibilities. To reduce regulatory risk and improve earnings predictability the period during which Light's existing tariff policy would remain unchanged was extended from five to eight years.

The implementation of some of the 1995 reforms may present difficulties due to their vagueness. This is the case, for instance, with the IPP law dealing with rate structure. The decree provides no guidance beyond a reference to general principles, and the law's provisions for energy sales to non-adjacent end-users are not elaborated. Such vagueness is perceived by many foreign investors as increasing the riskiness of investing in the sector. The Light concession has been criticized because it "...does not address the full range of return on investment issues typically considered in a US utility rate filing. Neither does it contemplate a completely free-market approach to general pricing consistent with the Chilean model. Consistent with the 1995 Concession Law, the government retains flexibility (after a period of agreed non-intervention) to make tariff adjustments throughout the term of the concession. Accordingly, foreign investors may justifiably question the reliability of long-term revenue projections developed for this and future offerings."(47)


Can the concession model of involving the private sector in public utilities be considered "privatization"? The rights of the concessionaire are limited: the property can revert to the state at the end of the concession period and the use of the property during the concession period is circumscribed by the concession contract itself and the regulator's interpretation of many of its clauses. Under such circumstances, why should the private domestic or foreign sector want to bid for a concession?

Private domestic and foreign groups are obviously attracted to a concessionary arrangement if they feel that the government is willing to interpret the contract in such a way as to make it possible to obtain a rate of return which is high enough to earn back all capital invested in the enterprise in the concession period and to make an internationally respectable profit. To date privatization has meant the inflow of capital to obtain concessions on already existing capacity. As it has been estimated that Brazil's investment needs in the electrical power sector in the period 1997-2000 amounts to over US$ 20 billion in order to avoid future shortages, and since the state will not have such resources at its disposal, reliance on the private sector, especially the foreign sector, should convince the government to make and interpret concessionary contracts in a way which is favorable to the investor. This is probably the reason why Brazil was successful in obtaining foreign participation in the privatization of Escelsa and Light, and why other federal and state firms in the sector should succeed in future privatizations via the concessionary method.(48)

There is always the possibility that the state will change its attitude again in the future. Once the pressing need for foreign capital in expanding the electric and other public utilities subsides, the foreign investor-friendly attitude of the 1990s may fade. A day may come when the political pendulum will once again swing towards redistributive populist attitudes which could once more result in much less favorable interpretations of concession contracts by the regulating agency. If on such occasions the foreign firms appeal to their governments for pressures to be brought on the host government for more favorable regulatory treatment, there might be a repetition of the type of confrontations which arose in the early 1960s.

From the point of view of the host government, the concessionary approach to running a public utility like electricity might carry with it certain advantages. The threat of losing a concession due to non-compliance with the contract or of not having a concession renewed upon expiration, gives the host government's regulatory agency some leverage over the concessionaire on such issues as efficiency, investment and technology.

Acknowledgment: We wish to thank Marina Figueira de Mello for many useful suggestions.


* Direct all correspondence to: Werner Baer, Department of Economics University of Illinois, 225 David Kinley Hall, 1407 W. Gregory Drive, Urbana, Il 61801. <>.

1. Five of these firms, however, were located in other sectors, such as EMBRAER (aircraft manufacturing), ESCELSA (electricity), MAFERSA (capital goods).

2. For a comprehensive review of issues surrounding regulation of public utilities, see: Pablo T. Spiller, "Institutions and Regulatory Commitment in Utilities' Privatization," Industrial and Corporate Change 2 (1993), pp. 287-450; Pablo T. Spiller, "How Not to Do It: Electricity Regulation in Argentina, Brazil, Chile, and Uruguay," University of California, Berkeley, Mimeo, 1992; and Brian Levy and Pablo T. Spiller (eds.), Regulations, Institutions, and Commitment: Comparative Studies of Telecommunications (Cambridge University Press, 1996).

3. Pierre Guislain and Michel Kerf, "Concessions - The Way to Privatize Infrastructure Sector Monopolies," in The International Forum for Utility Regulation and The World Bank Group, Public Policy for the Private Sector Infrastructure (Washington, D.C., 1996), p. 21.

4. See also Anthony W. Dries, "Franchising and Privatization," in Public Policy for the Private Sector Infrastructure, pp. 17-20; and Michael Klein and Neil Roger, "Back to the Future: The Potential in Infrastructure Privatisation," Silver Award Essay, The Amex Bank Review Prize Essays, American Express Bank, 1996.

5. According to Guislain and Kerf, France was a pioneer in granting concessions for the operation of infrastructure services by the private sector. In French concessions "...the state owns...(the)...assets from the moment they are built, but the private operator retains full control over them until the end of the concession period. In other cases...the legal ownership of assets built and financed by the private operator will remain private until their transfer to the state at the end of the concession term." Guislain and Kerf (1996), p. 22.

6. Ana Celia Castro, As Empresas Estrangeiras No Brasil, 1860-1913 (Rio de Janeiro: Zahar Editores, 1979), p. 83.

7. This large share was due to the ownership of the Light company by American Foreign and Power Company, which was Canadian-owned.

8. The best source on British investments in Brazil prior to World War I is: Richard Graham, Britain and the Onset of Modernization in Brazil, 1850-1914 (Cambridge University Press, 1968); data on foreign investments in Brazil in the 20th century can be found in: Werner Baer, The Brazilian Economy: Growth and Development, 4th edition (Westport, CT: Praeger, 1995), ch. 10.

9. Marcelo de Paiva Abreu, "British Business in Brazil, 1850-1950," Rio de Janeiro, Mimeo, 1996, p. 9.

10. In 1907 only 5% of energy used by industry consisted of electricity; by 1920 this percentage had risen to 47%. Mello (1996), p. 8.

11. Centro Da Memoria da Eletricidade no Brasil, Panorama do Setor de Energia Eletrica no Brasil (Rio de Janeiro, 1988), pp. 33-36.12.

12. Panorama, pp. 40-41.

13. Besides the Light group, foreign companies were also present in other places. For example, a British company, The Southern Brazil Electric Company, acquired controlling interest of the Empresa Eletrica de Piracicaba. Another British company obtained a concession to provide street lighting in the northeastern city of Recife.

14. Panorama, pp. 63-65. Two authors have stated that the reason for the small presence of domestic private capital in the sector "...was due in part to the scarcity of savings and the underdevelopment of capital markets...(and)...also to two other features of this period: the dominance of hydroelectricity, with its long lead times and high initial outlays, and low rates of return." Araujo and Besnosik (1992), p. 4.

15. In 1933 Guinle published an article in which he stated that the foreign power companies were earning " unjust remuneration, almost criminally so..." Politica, p. 16. In fact, since the mid-1920s there had been a notable press campaign against the dominance of foreign companies in the electric energy sector. See Politica, p. 19.

16. Although there was a provision for a 50 year concession in the case of a need for very large investments.

17. Panorama, pp. 80-83; Judith Tendler, Electric Power in Brazil: Entrepreneurship in the Public Sector (Cambridge, MA: Harvard University Press, 1968), pp. 48-49.18. Panorama, p. 91.

19. Also, the attacks of nationalists against the perceived privileges and monopolistic practices of foreign companies in public utilities during the debates in the constitutional assembly which was working on the 1946 Constitution, did not reassure foreign investors. Panorama, pp. 120-121. In the period 1930-45 the growth of electricity consumption in the states of Sao Paulo and Rio de Janeiro was 250%, while installed generation capacity only increased by 72%. It should also be noted that had foreign companies been willing to make substantial expansion, this would have been difficult during wartime conditions, due to restrictions on imports (especially capital goods).

20. For more details on Eletrobras, see Mello (1996).

21. Panorama (1988), pp. 142-143.

22. Politicas (1995), p. 64. It is interesting to note that in the 1954 National Electrification Plan there was a recognition that "...the legal rate of remuneration allowed for the electric sector was an obstacle to its expansion, as the credit market gave returns higher than 10%." The problem, however, could not be resolved by a simple increase of rates "...because the indirect effects on the foreign exchange situation would result in a significant increase of profit remittances by foreign firms." Politicas (1995), p. 64.

23. Panorama (1988), pp. 161-163.

24. Panorama (1968), pp. 164-170.

25. Dissatisfaction with the quality of power services in the early fifties resulted in a pronounced growth of self-generation, which reached 16.7% of installed capacity in 1953. Araujo and Besnosik (1993), p. 3.

26. Panorama (1988), p. 214.

27. In the early 1990s a number of government agencies had a hand in planning and control activities. Sector policies were made by the Ministry of Mines and Energy, which was advised in the 1980s by a National Energy Council and executed by Eletrobras. The National Department of Waters and Electric Energy (DNAEE) was in charge of regulation. For more detailed description, see: Araujo and Besnosik (1993), pp. 4-7.

28. Politicas (1995), pp. 94 and 101-103.

29. Baer (1995), pp. 112-116 and 259; see also Holanda Barbosa (1993), pp. 74-78.

30. Panorama (1988), pp. 224-225. Araujo and Besnosik mentioned that the 1980s an attempt was made to gradually introduce marginal cost pricing to replace average historical costs. But this was, in part, hindered by the inadequate amount of available information on cost structure. However, they claim that "...the introduction of marginal cost pricing was a welcome change from the prior empirical tariff setting methods. The change has been especially significant in view of the increasing trend in system expansion costs because of rising financial costs, of higher uncertainty over the expansion of the electricity marked market caused by prolonged economic crisis and of the need to install hydroelectric plants far from industrial centers. In particular, the system load curve has improved markedly thanks to a better balance between peak and off-peak tariffs." Araujo and Besnosik (1993), p. 9.

31. Araujo and Besnosik (1993), p. 9.

32. Baer (1995), ch. 6.

33. Politica (1995), p. 125.

34. It has been estimated that in the 1960s and 1970s investments in that sector often amounted to 2% of GDP. In the 1980s the Eletrobras group invested only about 0.8% of GDP and this declined to 0.4% of GDP in the period 1990-94. Looking at the Brazilian electric sector generation capacity in 1994 by date of investments, the following distribution emerged: Mello (1996, p. 9).

Installation before

1960: 9% 1961-70: 17% 1971-80: 48% 1981-89: 26% 100%

35. Mello points to the ironic fact that one of the reasons for the growth of the presence of the state in the electric sector in the 1950s and 1960s was the dearth of private investment funds, while the privatization of the 1990s was due to the dearth of official investment funds and the ample availability of resources from private domestic and foreign sources. Mello (1996), p. 11.

36. Concessoes (1995), p. 19.

37. Concessoes (1995), p. 21.

38. Mello (1996), p. 16.

39. Mello (1996), p. 10.

40. In fact, the consortium Iven already had previously acquired 20% of the shares of Escelsa from the state of Espirito Santo, and with a further acquisition of 27% of the shares became the principal shareholder. GTD acquired 25%. The federal government retained 16% of the shares, while the employees of the government obtained 7.7% and the state government of Espirito Santo 3.3%. See: Conjuntura Economica, Setembro 1995, pp. 34-35.

41. It was reported in Latin American Economy and Business of August 1995 that "...unlike most electricity companies, it has no debts; in fact at the end of last year it had US$ 280 million in cash a year and...assets of US$ 845 million. Analysts reckoned that the buyers had got a bargain. Before the auction 11 of July they reckoned that the company could fetch a premium between 20% and 25% over the minimum price" (p. 16).

42. In fact, the government continued to be the largest shareholder of Light. Adding the shares which were not offered and remaining in the hands of the government electricity holding company, Eletrobras, and the shares bought by BNDES (through its subsidiary BNDESPar), the government retained a total of 39.1% of Light's shares. See: Conjuntura Economica, Julho de 1996, p. 16. In late 1996 Goldman, Sachs, & Co. won the competition to coordinate the worldwide sale of 28.8% of Light shares which emained in the hands of Eletrobras.

43. According to one report, "CERJ is one of the least efficient electricity companies in the country: it loses 15% of its power to thieves and 12% of its power through its own technical shortcomings." Latin American Economy and Business, December 1996, p. 17.

44. One of the main tasks of the new owners of CERJ was to decrease the company's large energy losses, which were more than 28% (compared to the Brazilian average of 12%). The winning consortium had experience in transforming inefficient South American distribution companies into profitable ones. Between 1987 and 1995 Chilectra's system in Chile's central grid reduced losses on that system from 26% to 12%.45. As reported in the Jornal do Brasil, November 22, 1996.

46. For a more detailed discussion of changes in Brazil's tariff policies, see Mello (1996), pp. 14-16.

47. Project Finance, International, Yearbook 1997 (London: IFR Publishing, 1996), p. 58.

48. Upcoming privatizations include various subsidiaries of Eletrobras which in 1996 had a market capitalization of about US$ 16.2 billion. It will also include companies of individual states. The state of Sao Paulo plans to privatize: CESP, which is the largest electric generating company in Brazil, producing about 97% of the electricity generated in the state; CPFL, the fourth largest electric distribution company in Brazil; Eletropaulo, which is the largest distribution company in Latin America. The state intends to "unbundle" the existing industry and to set up separate businesses for generating, transmitting and distributing electric power - there would be 14 distribution companies, six generating companies and one transmission company. This strategy is expected to attract more buyers than a single block sale and to result in a competitive industry.


A Cerj e a Historia da Energia Eletrica no Rio de Janeiro. 1993. Rio de Janeiro: Centro da Memoria da Eletricidade no Brasil.

Araujo, Joao Lizardo de and Roberto I. Besnosik. 1993. Regulation, Institutional Structure and the Performance of the Brazilian Electricity Sector. Mimeo, Rio de Janeiro: Coppe/UFRJ.

Abreu, Marcelo de Paiva. 1996. "British Business in Brazil, 1850-1950," mimeo, Rio de Janeiro.

Baer, Werner. 1995. The Brazilian Economy: Growth and Development, 4th ed. Westport, CT: Praeger.

Castro, Ana Celia. 1979. As Empresas Estrangeiras No Brasil, 1860-1913. Rio de Janeiro: Zahar Editores.

Concessoes de Servicos Publicos no Brasil. 1995. Brasilia, DF: Presidencia da Republica.

Conjuntura Economica, various issues.

Dnes, Anthony W. 1996. "Franchising and Privatization." In Public Policy for the Private Sector Infrastructure. Washington, D.C.: The International Forum for Utility Regulation and The World Bank Group.

Graham, Richard. 1968. Britain and the Onset of Modernization in Brazil, 1850-1914. New York: Cambridge University Press.

Guislain, Pierre and Michel Kerf. 1996. "Concessions - The Way to Privatize Infrastructure Sector Monopolies." In Public Policy for the Private Sector Infrastructure. Washington, D.C.: The International Forum for Utility Regulation and The World Bank Group.

Historia e Energia 6: O capital privado na reestruturacao do setor eletrico Brasileiro. 1996. Sao Paulo: Eletropaulo, Departamento de Patrimonio Historico.

Holanda Barbosa, Fernando de. 1993. "The Efficiency of State Intervention in the Economy." In Brazil and the Challenge of Economic Reform, edited by Werner Baer and Joseph S. Tulchin. Washington, D.C.: The Woodrow Wilson Center Press; distributed by The Johns Hopkins Press.

Klein, Michael and Neil Roger. 1996. "Back to the Future: The Potential in Infrastructure Privatization," Silver Award Essay, The Amex Bank Review Prize Essays, American Express Bank.

Levy, Brian and Pablo T. Spiller (Eds.). 1996. Regulations, Institutions, and Commitment: Comparative Studies of Telecommunications. New York: Cambridge University Press.

Mello, Marina Figueira de. 1996. "Os impasses da privatizacao do setor eletrico," PUC/RIO, Departamento de Economia, Texto Para Discussao No. 365, Dezembro.

Panorama do Setor de Energia Eletrica no Brasil. 1988. Rio de Janeiro: Centro da Memoria da Eletricidade no Brasil.

Politicas De Governo e Desenvolvimento Do Setor de Energia Eletrica: Do Codigo de Aguas a Crise dos Anos 80 (1934-1984). 1995. Rio de Janeiro: Centro da Memoria da Eletricidade no Brasil.

Spiller, Pablo T. 1993. "Institutions and Regulatory Commitments in Utilities' Privatization." Industrial and Corporate Change, 2.

Spiller, Pablo T. 1992. How Not to Do It: Electricity Regulation in Argentina, Brazil, and Uruguay. Mimeo, University of California, Berkeley.

Tendler, Judith. 1968. Electric Power in Brazil: Entrepreneurship in the Public Sector. Cambridge, MA: Harvard University Press.
COPYRIGHT 1998 JAI Press, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:The Changing Role of International Capital in Latin America
Author:Baer, Werner; McDonald, Curt
Publication:Quarterly Review of Economics and Finance
Date:Sep 22, 1998
Previous Article:Private investment response to neoliberal reforms in a delegative democracy: reflections on Argentina.
Next Article:Beyond privatization: getting the rules right in Latin America's regulatory environment.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters