Brazil’s Import-Substitution Industrialization

Oxford Handbooks Online

Brazil’s Import-Substitution Industrialization
Werner Baer
The Oxford Handbook of the Brazilian Economy
Edited by Edmund Amann, Carlos R. Azzoni, and Werner Baer

Print Publication Date: Oct 2018

Subject: Economics and Finance, Urban, Rural, and Regional Economics, Industrial
Organization
Online Publication Date: Aug 2018 DOI: 10.1093/oxfordhb/9780190499983.013.5

Abstract and Keywords

This chapter examines the development of Brazil’s inward-oriented industrialization

strategy, commonly termed “import-substitution industrialization” (ISI). Originating in the
1930s under the corporatist administration of Getúlio Vargas, by the 1960s the strategy
had succeeded in transforming the structure of the Brazilian economy, turning it into a
major industrial powerhouse. Successful though the strategy initially was in promoting
growth and structural change, it nevertheless suffered from inherent flaws, notably its
heavy reliance on imported inputs and a failure to produce and export efficient industrial
sector. This chapter considers the achievements and failings of ISI in some detail and also
discusses the results of attempts to reintroduce the strategy on a limited scale in the first
decade of the 2000s.

Keywords: industrialization, Brazil, industry, import substitution, export

5.1. Early Industrial Growth

SUBSTANTIAL industrial growth in Brazil began in the late nineteenth century. It was
associated with the large wave of European immigration that followed the abolition of
slavery in 1888. Free immigrant labor represented a large enough market to lead many
importers of manufactured goods (mainly textile and food products) to establish local
production facilities. By 1907 there existed close to 3,000 manufacturing establishments,
employing about 136,000 workers in such sectors as textiles, food processing, and
beverages. Capital goods production did not exist, except for some primitive ironworks,

Page 1 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

naval yards, and repair shops for the transportation sector (Dean 1969, Chapter 6; Villela
2011, 41).

World War I caused a shortage of many types of manufactured goods, as industrial

countries switched from civilian to war goods production. The shortage of manufactured
imports did not result in investments in new capacity, as Brazil did not have a capital
goods industry. The shortages were partially met by increased use of existing production
capacity (Baer 2014).

The dynamism of Brazil’s economy in the 1920s was based on a booming coffee sector, as
its share of exports rose from 56% in 1919 to 75% in 1924. A slight appreciation of the
exchange rate and rising internal prices decreased the protection of domestic industry
from foreign competition, and throughout most of the 1920s industry grew at a very slow
pace. The average yearly growth rate of industrial output fell from 4.6% in 1911–1920 to
3% in 1920–1929. Since the textiles sector constituted the principal industrial sector at
the time, its stagnation explains the overall weak performance of industry. A closer
(p. 90) examination, however, reveals much faster growth in other industrial subsectors,

such as metallurgical products, cement, iron and steel, and paper products (Baer 2014,
34–35; Villela 2011, 42).

Despite the growth of industries since the 1890s, one cannot define Brazil’s industrial
growth as “industrialization.” That term would only apply where industry becomes the
leading growth sector of the economy.

5.2. Import Substitution of the 1930s

The depression of the 1930s had a severe negative impact on Brazil’s exports, whose
value fell from US$445.9 million in 1929 to US$180.6 million in 1932. In addition to the
decline in export receipts, the inflow of foreign investments had come almost to a
complete halt. This situation became critical as the foreign exchange needed to finance
the country’s external debt (which amounted to US$1.3 billion in 1931), not counting the
remittances of profits by private entities, forced the government to take drastic actions
(Abreu 1990, Chapter 3, for further details).

The government suspended part of its foreign debt services and introduced exchange and
other direct controls, combined with a devaluation of the currency. The latter increased
the price of imports. The combination of these measures resulted in a decline of imports
from US$416.6 million in 1929 to US$108.1 million in 1932.

Additionally, with the steep decline in world coffee prices, the federal government took
over the coffee-support program from individual coffee-producing states (mainly São
Paulo). A National Coffee Council was founded, which bought all coffee production,

Page 2 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

destroying a large quantity that could not be sold or stored (for further details, see Baer
2014, 38; Dean 1969, Chapter 7; Furtado 1972).

The curtailment of imports and the continued domestic demand resulting from the
income generated by the coffee support program caused shortages of many
manufactured goods and a consequent rise in their relative prices. This acted as a
catalyst for a burst of domestic industrial production. By 1931 industrial production had
fully recovered from a decline that had begun in 1928, and it doubled in the following
eight years. Especially noteworthy was the growth of textile production (147%), metal
products (three times larger in the late 1930s than in 1929), and paper products (almost
seven times larger).1

Unlike during World War I, industrial growth was based not only on the more intensive
use of existing production capacity, but also, especially in the second half of the 1930s, on
the creation of new capacity. One could characterize the rapid industrial growth of the
1930s as Brazil’s first experience with import-substitution industrialization, that is, the
industrial sector became the Brazilian economy’s leading sector, whereas in earlier years
industrial growth had accompanied the expansion of the primary export sector.

5.3. Import-Substitution Industrialization

(p. 91)

as a Deliberate Development Policy

During World War II there was substantial industrial growth, but with little expansion of
productive capacity. Industrial production grew at an annual rate of 5.4% in the period
1939–1945. Noteworthy in this period were average yearly growth rates of metal
products (9.1%), textiles (6.2%), shoes (7.8%), and beverages and tobacco (7.6%), which
were industries whose imports were drastically curtailed. Investment activities fell at
first, but rose again in 1945. This was mainly due to the capital equipment Brazil was
allowed to import during the war to construct its first large integrated steel mill at Volta
Redonda (Baer 1969).

The establishment of the Volta Redonda plant is especially significant in that it marked
the most direct participation of the state thus far in productive activities. The
construction and operation of the plant occurred under the auspices of a state-owned
enterprise, the Companhia Siderurgica Nacional (CSN; National Steel Company). The
expanding role of the state in the industrialization process and, indeed, with the process
of development more generally formed part of the corporatist ethos associated with
President Vargas’s Estado Novo (New State). This marked a radical change from the less
interventionist approach associated with successive administrations during the
nineteenth and early twentieth centuries.

Page 3 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

The drastic decline of imports during World War II and the boom of exports resulted in a
substantial increase in the country’s foreign exchange reserves, from US$71 million prior
to the start of the war to US$708 million in 1945. In February 1945, the government
established a foreign exchange regime without restrictions, except for some limitation on
the remittance of profits. Brazil’s currency, the cruzeiro, was kept at its prewar value of
CR$18.50 per dollar and did not change until 1953, although prices rose 285% from 1945
to 1953.

The combination of an overvalued currency and the elimination of most import

restrictions resulted in a spurt of imports, which eliminated the accumulated wartime
foreign exchange reserves within one year. This forced the government to reintroduce
foreign exchange controls, which were also supplemented by import controls through a
system of import licensing (Baer 2014, 54–61).

The post–World War II industrialization drive was initially the consequence of measures
taken in order to cope with balance of payments difficulties. Only gradually,
predominantly in the 1950s, did various efforts at balance of payments protection become
part of a deliberate policy to promote industrialization through import substitution. The
protection of domestic industry occurred through different systems of exchange controls
and in some years through a multiple exchange rate system (Baer 2014, 54–61).

The government also made use of protective tariffs and of SUMOC2 Instruction
(p. 92)

113, which was designed to attract foreign investments by enabling firms to import
capital equipment without the need for exchange cover. The Tariff Law of 1957 expanded
and solidified protection of domestic industry. In many cases tariffs were as high as 60%,
80%, and 150%.

The creation in December 1950 of the Joint Brazil–United States Economic Development
Commission provided an important vision to policymakers about infrastructural
bottlenecks and the potential for structural changes in the economy. The Commission also
recommended the creation of a development bank, which was established in 1952 as the
BNDE (National Bank for Economic Development).3 It was to become the basic financier
of state and private enterprises. Its role became especially significant in financing heavy
capital investment across industry as the industrialization process intensified into the
1960s and the 1970s.

Import-substitution industrialization (ISI) resulting from these measures had a notable

growth impact in the first half of the 1950s (with industrial production growing at 9% per
year), but this expansion would be even more pronounced in the second half of the 1950s
under the presidency of Juscelino Kubitschek, who established a Targets Plan, which
made the state even more active in speeding up ISI. Under this plan, specific programs
were established to promote such industries as automobile and utility vehicle
construction, shipbuilding, and heavy machinery. The programs were organized through
the BNDE, and the favored industries were given special treatment for importing capital

Page 4 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

goods, raw materials, components, and the like for specific periods of time (Baer 2014;
Villela 2011).

Brazil’s ISI program did not rely solely on foreign private investments; it was also
complemented by a substantial growth in state enterprises. The establishment of large
integrated steel mills was deemed essential to the deepening of the industrial sector, and
since there was little interest from the foreign private sector and the domestic private
sector did not have the technical and financial capabilities to enter this field, state firms
were established (Baer 1969). Additionally, the expansion of public utility services, such
as power generation and distribution, was increasingly provided by public firms (both at
the federal and state levels), as private domestic and foreign firms showed little interest
in an increasingly regulated sector.

5.4. The Impact of Import-Substitution

Industrialization Policies
The ISI policies of the 1950s resulted in high rates of economic growth. The average
yearly real growth rate of gross domestic product (GDP) in the period 1947–1962 was
over 6%, and in the years 1956–1962 it was 7.8%. From 1947 to 1961 the overall real
product increased by 128%. The real agricultural product rose by only 87%, while the
(p. 93) industrial product increased by 262%. For the absolute increase of GDP between

1947 and 1961, agriculture was responsible for only 18%, whereas the nonagricultural
sector contributed the rest.

Page 5 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

Table 5.1 Brazil: Sectoral Distribution of GDP (Percentages)

1950 1960 1970 1980 1990 2010

Agriculture 13.3 10.2 4.6 5.0 5.8 6.2

Industry 28.7 32.2 33.0 29.9 32.4 30.8

Services 58 57.6 62.4 65.0 61.8 63.1

100 100 100 100 100 100

Source: Conjuntura Econômica; IBGE, Sistema de Contas Nacionais.

Page 6 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an
individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

Page 7 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

Table 5.2 Brazil: Sectoral Distribution of Labor (Percentages)

1950 1960 1980 1990 2000 2010

Agriculture 64.4 58.9 38.1 26.4 22.3 16.7

Industry 16.5 16.8 22.8 22.8 19.5 20.3

(Manufacturing (11.5) (11.8) (12.7) (14.7) (12.0) (12.1)

)

Services 19.1 24.3 39.1 50.8 58.2 63.0

100.0 100.0 100.0 100.0 100.0 100.0

Source: Conjuntura Econômica; IBGE, Anuário Estatístico do Brasil, various years.

Page 8 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an
individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

From the 1930s on, a characteristic of Brazil was pronounced rural-urban migration. In
the early post–World War II years, the rate of the country’s urban population growth was
almost twice that of the general population. Unfortunately, as is clear from the data in
Tables 5.1 and 5.2, the industrial sector absorbed only a small proportion of migrants to
the cities. Most survived in the service sector, which at the time consisted to a large
extent of marginal activities, including informal street vending.

The lack of labor absorption by the most dynamic sector was a puzzle that generated a
fairly large literature trying to explain the phenomenon. Some claimed that there was a
distortion in factor prices (industrial wages made artificially higher through social
legislation than would be warranted, given the huge supply of labor), while capital was
made artificially cheap by the development bank and other incentives to invest in import-
substitution industries. Assuming different production techniques to choose from, this
distortion of factor prices led to a tendency to choose the more capital-intensive ones.
However, although this may offer a logical explanation, the real world was rather more
complicated. Many multinationals that invested in Brazil chose to ship older capital
equipment to their Brazilian plants (which was possible through special Brazilian (p. 94)
import regulations); thus Brazil’s ISI in the 1950s and 1960s utilized comparatively labor-
intensive equipment in relation to the technological state of the art at the time. In some
other industries, such as special types of steel, production technology was rigid and was
not influenced by relative factor prices.

The capital-intensive nature of Brazil’s new industries contributed to a continued

concentration of income. In other words, as the capital/labor ratio of the dynamic sectors
of the economy (i.e., industry) was greater than the capital/labor ratio of the traditional
sectors, the net result (other things being equal) was an increase in the concentration of
income in Brazil.

5.5. The Import Coefficient Problem

Brazil’s government promoted ISI across the board. Its major goal was to lower the
import coefficient that is, the lowering of imports as a share of GDP. The value of imports
of goods and services as a proportion of GDP declined from 9% in 1949 to 5% in 1960.4

During this period of import substitution, exports were neglected and little attempt was
made to diversify them. In 1955, coffee still accounted for 59% of exports, and by 1964,
53%; other primary products made up 38% in 1955 and 45% in 1964. However, by the
1960s, imports as a share of GDP began to rise again, reflecting a rise of imported inputs
into industry (Baer 2014, 193–194). It was at this stage that policymakers realized that
there was a limit to ISI and that export diversification was crucial in order to assure
future foreign exchange earnings to pay for imported inputs into Brazil’s industries. The
problem, however, was that most ISI industries were established behind high tariff walls,

Page 9 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

with high costs and many firms producing goods of low quality.5 It would take special
incentives (such as lower taxes for export earnings and subsidized credit for export) to
begin a process of export diversification.

5.6. Regional Inequalities

Inequalities in the geographic distribution of income and growth have been a
characteristic of the Brazilian economy since colonial times. By the time of ISI, Brazil’s
major export product was coffee, and most of its production was located in the Southeast
of the country, while only 43% of the population lived in that region. Most of Brazil’s new
industries were located in the Southeast, and by the early 1960s this region accounted for
more than 75% of the country’s industrial product (Baer 2014, Chapter 12). Most new
industries were located in the states of São Paulo, Rio de Janeiro, and Minas Gerais. The
explanation for this regional concentration was that these states had the highest per
capita income and thus represented the greatest source of demand for industrial
products. They also had the best infrastructure and most of the skilled labor.

(p. 95) The concentration of ISI in the Southeast of the country resulted in a dynamic that
worsened the situation of the poorer regions, especially Brazil’s Northeast, which at the
time had more than a third of the country’s population. The Northeast continued to
produce primary export products (such as sugar, cotton, and cacao), but with ISI was
forced to purchase its manufactured products from Brazil’s Southeast, whose prices were
much higher than imported products. Thus the terms of trade of the Northeast declined,
which meant that the poorest region of Brazil was transferring resources to the most
prosperous, the Southeast. In other words, the analysis of the center-periphery
relationship developed by the original theorists of ISI—Raúl Prebisch and his Economic
Commission for Latin America (ECLA) group—and which was used to justify ISI, now
repeated itself within Brazil (Baer 2014, 254–262).

Page 10 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

5.7. Neglect of Agriculture

Brazil’s investment in agriculture during the intensive ISI period was quite small,
especially in agricultural products for the domestic market. Although the growth of the
population was smaller than the growth of the food supply, there was another factor that
cast a shadow on this favorable picture. While the intensive rural-urban migration
resulted in an urban population growth of about 5.4% per year in the 1950s, most of the
increase of food production occurred in new lands placed under cultivation, rather than
increased productivity in existing lands. Since the rapidly rising demand for food in urban
centers had to be supplied from increasingly distant places, there was an increasing
strain on the country’s precarious rural-urban transportation network and on the
agricultural marketing system. It was generally recognized by the 1960s that further
industrial growth would be severely hampered if advances were not made in agricultural
productivity near the main consuming centers. If these trends continued, the rise of
relative food prices would not only increase inflationary pressures, but also lead to rising
social tensions.

Page 11 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

5.8. The 1960s: Import-Substitution

Industrialization in Crisis
By the early 1960s the Brazilian economy had lost its dynamism. After GDP growth
reached a peak of 10.3% in 1961, it fell to 5.1%, 1.5%, and then 2.4% in 1962, 1963, and
1964, respectively. The immediate cause of the stagnation that set in after 1961 was the
continuing political crisis that the country experienced after the resignation of Jânio
Quadros from the presidency in August 1961. The turbulent years that followed until the
overthrow of the Goulart government in April 1964 were devoid of any consistent
economic policy.6

(p. 96) After the coup of 1964, the new military regime concluded that the path to
economic recovery lay in remedying a number of problems that had emerged during the
ISI period. The primary concerns at first were with bringing inflation and its price
distortions under control, modernizing capital markets, indexing controlled prices, and
using tax incentives to influence the allocation of resources in desired directions (such as
investments in neglected regions) (Baer 2014, 73–74).

Foreign trade policy was considered of central importance by the post-1964 military
governments. The rapid growth and diversification of exports was deemed essential to
the long-term health of the economy. In order to achieve this diversity, state export taxes
were abolished, administrative procedures for exporters were simplified, and export tax
incentives and subsidized credit were instituted. In 1968 a crawling-peg exchange rate
policy was introduced to avoid overvaluation of the currency. It consisted of frequent (but
unpredictable) small devaluations of the cruzeiro. This would keep the currency from
becoming overvalued as long as some inflation was still present, while keeping
speculation against the currency at a minimum and keeping the exchange rate from
becoming a political issue.

The outward orientation of policies on the import side consisted mainly of a tariff reform
in 1966, which resulted in the lowering of nominal tariffs from an average of 54% in
1964–1966 to 39% in 1967. Subsequent changes again led to a rise in rates, but not to
pre-reform levels. Real protection was also reduced in the late 1960s and early 1970s by
the fact that the rate of devaluation of the cruzeiro was smaller than the rate of inflation.
Collectively, these policies, which pushed the industrialization strategy in a more
outward-oriented direction, have become known as post-ISI. For a while this policy set
proved highly effective, giving rise to a period from the late 1960s to the early 1970s
known as the “Brazilian Economic Miracle.”

After a period of stagnation (1962–1967), when annual real GDP growth was only 3.7%,
Brazil experienced a period of high growth that lasted from 1968 to 1973, when the
annual growth rate of GDP averaged 11.3%, which was generally attributed to the
reforms instituted by the military regime. During those years, industry was again the

Page 12 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

leading sector, expanding at yearly rates of 12.6%. Within manufacturing, the highest
growth rates were achieved in transport equipment, machinery, and electrical equipment,
while traditional sectors, like textiles, clothing, and food products, experienced much
slower growth rates.

External trade grew at rates substantially higher than the economy. In the years 1970–
1973, the average yearly growth rate of exports was 14.7% and of imports 21%. The
resulting trade deficit was also accompanied by a rising deficit in the service balance.
Until 1974, however, this was more than covered by a massive inflow of official and
private capital.

In these years Brazil succeeded in diversifying its commodity export structure, and on the
import side there was a notable increase in capital goods. Coffee, which in 1964
amounted to 53% of exports, fell to 13% in 1974; manufactured goods rose from 5% to
36%; and capital and intermediate goods rose from 60% of imports in the early 1960s to
85% in 1972. The post-1964 policies clearly opened the economy (p. 97) to foreign trade.
Whereas the ISI policies of the 1950s decreased the import coefficient (import/GDP ratio)
from 16% (1947–1949) to 5.4% in 1964, it rose again to 14% in 1974.

5.9. The Oil Shock of 1973 and Its

Consequences
The oil shock of November 1973 quadrupled the price of petroleum. At the time Brazil
was relying on imports for over 80% of its oil consumption and thus its import bill rose
from US$6.2 billion in 1973 to US$12.6 billion in 1974, and the current account from a
deficit of US$1.7 billion to US$7.1 billion.

At the time Brazil had two options for reacting to the oil shock: it could either
substantially reduce growth in order to diminish its non-oil import bill, or it could opt for
continued relatively high growth rates. The latter would cause a substantial decline in the
country’s foreign exchange reserves and/or a substantial increase in its foreign debt. It
opted for the latter.

In 1975 the Second National Development Plan (PND II) was introduced. It consisted of a
huge investment program, with the following two goals: (1) import substitution of basic
industrial products (such as steel, aluminum, copper, fertilizers, and petrochemicals), and
capital goods;7 and (2) the rapid expansion of economic infrastructure (hydro and nuclear
power, alcohol production, transportation, and communications). Many of these
investments were undertaken by state enterprises (in energy and steel), whereas others
(especially capital goods) were carried out by the private sector, with massive financial
support from the BNDES. The goals of these programs were (1) to act as a strong
countercyclical policy vis-à-vis the impact of the oil crisis and maintain a reasonable rate

Page 13 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

of growth; (2) to change the structure of the economy through a new round of import
substitution and export diversification and expansion; and (3) to encourage international
lenders to finance the current account deficit and to postpone external adjustment.

It has also been claimed that the basic ideas behind PND II were to increase the country’s
self-sufficiency in sectors such as energy and to develop new types of comparative
advantages. Large sums of state investments that occurred at the time were justified on
the grounds that in the short run the returns on investment in infrastructure and heavy
industry would be too low to attract private capital.

The growth impact of the PND II was quite positive, as the yearly real GDP expansion was
7% (led by industry, which grew at yearly rates of 7.5%). The sectors that experienced
exceptional growth rates during this period were metal products, machinery, electrical
machinery, paper products, and chemicals (Baer 2014, 78). Taking the ratio of imports to
domestic production as a measure of ISI, there occurred a notable decline between 1976
and 1981 in following sectors listed in Table 5.3. (p. 98)

Table 5.3 Import Coefficient, Selected Sectors 1976 and 1981

1976 1981

Paper 0.13 0.08

Cellulose 0.05 0.01

Polyethylene 0.72 0.02

Plastic tubes 0.45 0.03

Steel 0.15 0.05

Fertilizers 1.34 0.85

Aluminum 0.58 0.14

Capital goods 0.64 0.40

Brazil’s option of growth through external debt was justified on the grounds that future
savings of foreign exchange resulting from the investment programs—due to import
substitution and to the development of new export capacity—would ultimately bring
about a situation in which Brazil could produce trade surpluses large enough to service
and repay its international debt.

Page 14 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

5.10. The Debt Crisis and Its Aftermath

The debt crisis, which exploded in the early 1980s and resulted in the (economically) “lost
decade” of the 1980s, led the country to push hard to promote nontraditional exports and
to decrease imports. Throughout the 1980s, Brazil had to negotiate with its foreign
creditors and the International Monetary Fund (IMF) to turn over and extend the
servicing of its debt. The conditions for various adjustments of the foreign debt gradually
forced Brazil into a full-scale acceptance of what become known as neoliberal policies (for
a discussion of neoliberalism in Brazil, see Amann and Baer 2002).

Tariffs were gradually lowered, the market reserves of various products (such as
computers) were eliminated, many state-owned industries were privatized, and various
artificial stimuli for exports were removed (Silber 2011). The first round of cautious trade
reforms took place under the civilian administration of President Sarney in 1987. Then, in
1990, the directly elected president Fernando Collor de Melo instituted a rolling four-year
program of trade liberalization that became known as the Abertura Comercial (Trade
Opening). This comprised phased tariff reductions and, more importantly, a
comprehensive rolling back of nontariff barriers. The latter, including outright import
prohibitions (the so-called Anexo C) had comprised the main protectionist mechanisms of
the ISI era.

Besides Brazil’s unilateral trade reform efforts, the early 1990s saw the creation of
MERCOSUL, a customs union embracing Brazil, Argentina, Paraguay, and Uruguay.
(p. 99) Suddenly, regional trade in industrial products became significantly freer. The

partial exception to this was the still politically sensitive automobile sector. In the latter, a
managed trade regime was established between MERCOSUL’s two major automotive
producers, Brazil and Argentina.

In addition, various measures were instituted to facilitate foreign investments. These

centered on privatization of public utilities and state-owned enterprises, among them the
poster child of the early ISI era, CSN. Another icon of Brazil’s industrialization drive,
Petrobrás, saw its monopoly in domestic onshore and offshore oil exploration activities
broken as the result of a 1995 constitutional amendment. Collectively, the intent of these
policies was to increase efficiency through foreign competition and to increase the inflow
of foreign direct investment.

Economic liberalization gained momentum throughout most of the 1990s, and, as Villela
(2011, 53) observes, “[. . .] stabilization, trade liberalization and privatization combined to
usher in a new development model, leaving behind 60 years or so of ISI. Globalization
had finally caught up with Brazil, which embraced market-friendly reforms more out of
necessity than from actual conviction as to their merits.”

Page 15 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

5.11. Inflation Targeting, Interest, and the

Exchange Rate
The success of Brazil’s Plano Real (Real Plan) stabilization program (introduced in 1994)
was in part connected with the appreciation of Brazil’s currency, which was the result of
the high interest rates that were important stabilization instruments used at the time.
These high base interest rates (varying between 7% and 10% in real terms) attracted a
substantial amount of capital, which appreciated the currency and thus facilitated
stabilization efforts. For a while this came to be known as a policy of controlling inflation
via an exchange rate anchor. Although this policy was discontinued at the turn of the
century and a policy of inflation targeting was introduced, there has been a continued
emphasis on the use of high interest rates to underpin price stability. As during the first
decade of the twenty-first century Brazil’s interest rates were among the highest in the
world and attracted substantial portfolio inflows, Brazil’s real came to be one of the most
appreciated currencies in the world.

The strong real made it very difficult for many of the country’s industries to achieve
international competitiveness. However, this did not result in balance of payments
problems since at the same time there occurred for over a decade a world commodity
boom, originating in the high growth rates of a number of Asian countries, led by China.
Thus over the second half of the first decade of the 2000s commodities rose to over 60%
of total exports, while manufactured goods declined to 36%. This is ironic, since what
originally led Brazil and similar countries to industrialize was too great a dependence on
the export of commodities. By the second decade of the twenty-first century, however,
commodity prices declined and contributed to a period of stagnation.

It is not clear to what extent Brazil’s appreciated exchange rate was responsible
(p. 100)

for the decline of manufactured exports. It could be that Brazil’s use of tax incentives and
subsidized credit to stimulate nontraditional exports led some of its trade partners to
interpret these policies as amounting to dumping and thus to take retaliatory action.

Page 16 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

5.12. Import-Substitution Industrialization

Redux? Industrial Policy under Presidents Lula
and Rousseff
As a partial response to the deindustrialization process that had been unfolding since the
early 1990s, the PT (Workers’ Party) administrations of President Lula and President
Rousseff (2002–2016) gradually adopted a more interventionist industrial policy stance.
To some extent, this development can also be seen as a product of the growing influence
of the structuralist school of economic thought within the administration. One aspect of
the resurgence of interventionism centered on an expansion of BNDES lending to
strategic or technologically dynamic sectors, notably biotechnology, aerospace, and
automotive products.

This development can, in many ways, be seen as merely an intensification of a long-

established facet of public policy in Brazil; even the more free-market-oriented
administration of Fernando Henrique Cardoso (1995–2002) had engaged in reasonably
activist industrial and technology policy. What made the policy stance of his two
successors more eye-catching was their willingness to revert to protectionist measures
when deemed necessary. Thus in the rapidly expanding oil and gas sector, the leading
national oil major, Petrobrás, was obliged to purchase equipment from domestic
producers. As a result, there was a rapid expansion in the shipbuilding and offshore
engineering sector. The continued growth of this sector looked set to be all but
guaranteed by the program to develop newly discovered “pre-salt” offshore oil deposits.

While the ISI era might not be returning, a sense that the cause of trade liberalization
was, at least, in retreat was further evidenced by the lack of progress in resuming tariff
reductions or signing fresh free trade agreements. By the end of the Rousseff
administration in 2016, tariffs on industrial products in Brazil remained very high by
international standards. In the automotive sector, especially, the trade regime made it all
but commercially impossible to import assembled vehicles. Thus, as during the 1960s and
1970s, most domestic demand was met through Brazil-based assembly plants owned by
foreign multinationals.

Looking to the future, what prospects remain for the ISI-style mechanisms adopted by
Lula and Rousseff? Following the arrival of a more centrist administration in mid-2016,
the indications so far point toward the pursuit of a more pro-free-trade agenda, in
(p. 101) particular involving the signing of trade accords with regional and extra-regional

partners (including, possibly, the post-Brexit UK). At the same time, the role of the
BNDES has already been scaled down and lines of credit cut back.

Perhaps most significant in political terms, the entire framework surrounding the
Petrobrás procurement process is the subject of a wide-ranging criminal investigation.
This investigation—known as Lava Jato—has already highlighted the means by which an
Page 17 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

activist trade and industrial policy offered an avenue for the corrupt financing of political
parties. It also provided one of the central de facto motives for the impeachment of
President Rousseff. With Lava Jato ongoing, and oil prices well below their 2008 peak,
investment in the oil sector has sharply contracted. This has in turn resulted in a financial
crisis in Rio de Janeiro state, home to most of the country’s oil services and equipment
activities. Thus, as Brazil approaches the third decade of the twenty-first century, it is fair
to say that the ISI policy agenda is well and truly again in retreat.

5.13. The Infant That Didn’t Grow?

Many apologists for ISI in Brazil (and other countries that followed its model) have
claimed that ISI was an application of the traditional “infant industry” argument for
protection, that is, that in a closed economy many new industrial sectors would have the
opportunity and time to “learn by doing,” and after a while many protected industries
would achieve productivity levels that would make them internationally competitive.

Unfortunately, extensive studies carried out under the sponsorship of IPEA (the research
center of Brazil’s Planning Ministry) found that many sectors of Brazil’s industry were
substantially lagging behind in productivity (De Negri and Cavalcante 2014). It was noted
that by the end of the first decade of the twenty-first century, Brazil’s labor productivity
was about 25% of productivity in rich countries (De Negri and Cavalcante 2014, 37–38).
It was also shown that in 1995 the productivity of the United States was 6.6 times greater
than that of Brazil, and by 2009 had grown to 7.1 times Brazil’s productivity.

The IPEA productivity studies place the blame of Brazil’s low productivity on the
country’s low human capital, low research and development (R&D) as a proportion of
GDP, and decades of low investment in the country’s infrastructure.

5.14. The End of Import-Substitution

Industrialization and Deindustrialization
In the last decade, many studies of emerging countries have noted the phenomenon of
deindustrialization, that is, shifting toward services at an earlier stage of development
(p. 102) than was the case in older industrial economies.8 This has also been the case in

Brazil (see Table 5.1). Between 1950 and 1980, agriculture’s share declined from 13.3%
to 4.6%, industry’s rose from 28.7% to 33%, while services’s share changed very little,
from 58% to 62.4%. In roughly the same period, from 1950 to 1980, one can observe a
change in the sectoral distribution of employment (Table 5.2); agriculture declined from
64.4% to 38.1%, industry rose from 16.5% to 22.8%, and services from 19.2% to 39%.

Page 18 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

By 1980 the share of industry in GDP reached its peak of 33% (with manufacturing
reaching 21.1%), and similarly, employment in industry reached 22.8% (with
manufacturing reaching 12.7%). Bonelli et al. (2013, 69) analyzing Brazil’s
industrialization via shares of GDP in constant prices, find that this process began in the
mid-1970s and continued steadily until the second decade of the twenty-first century.
Within manufacturing they note a decline of the shares of textiles and related products,
chemical products, and plastics, while there was a notable growth in the shares of
pharmaceutical products, machinery, electrical equipment, and transport equipment.

In the mid-1990s, as already noted, Brazil introduced a successful stabilization program

while the country followed policies of trade and market liberalization. Using interest rates
as one of its principal policy instruments and adopting a floating exchange rate, there
occurred a long period of currency appreciation. This has had a negative effect on
Brazilian industry, making it less internationally competitive and contributing to a decline
in its participation in GDP, falling to 18% by 2010.

Traditionally, industries have been viewed as the main driver of economic development,
while services have been associated with low-productivity activities (Baumol 1967). The
best-known exposition of this is found in the writings of Kaldor (e.g., 1966), who viewed
manufacturing as the leading “engine of growth.” However, recent experiences, such as
that of Brazil, present a slightly different picture. Growth has gone hand in hand with an
increasing share of the tertiary sector in total GDP and employment. This shift has been
called “tertiarization,” and seems to be occurring at an earlier stage of economic
development than was the case with older industrial economies. It has in turn prompted
fears of “premature deindustrialization” (McMillan and Rodrik 2011).

The latter has to take into account that services do not automatically close avenues to
industrial development. There has been an increasing trend in the twenty-first century for
industry to outsource tasks that were previously performed internally. Thus the growing
role of services does not automatically imply deindustrialization.

Also to be taken into account is that “modern” services can create positive spillovers into
the entire economy. Many modern services have become increasingly tradable, highly
productive, and technologically intensive (Rowthorn and Ramaswamy 1999). As far as
Brazil is concerned, evidence suggests that structural change has favored low-
productivity activities within services and that structural change has not been growth-
enhancing (Cruz et al. 2008; McMillan and Rodrik 2011). However, it must be noted that
the quality of employment has increased since 2000. Within services, labor has shifted
from informal to formal activities. The share of labor in formal services has increased
from 48% to 55% between 2000 and 2009 (Aldrighi and Colistete 2012). When viewed
(p. 103) from this perspective, Brazilian services may not have performed as badly as has

been suggested.

Page 19 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

References
Abreu, Marcelo de Paiva. 1990. A ordem do progresso: Cem anos de política econômica
republicana, 1889–1989. Rio de Janeiro: Editora Campus.

Amann, Edmund. 2000. Economic Liberalization and Industrial Performance in Brazil.

Oxford: Oxford University Press.

Amann, Edmund, and Werner Baer. 2002. “Neoliberalism and Its Consequences in Brazil.”
Journal of Latin American Studies 34 (4): 945–959.

Baer, Werner. 1969. The Development of the Brazilian Steel Industry. Nashville, TN:
Vanderbilt University Press.

Baer, Werner. 2014. The Brazilian Economy: Growth and Development, 7th edition.
Boulder, CO: Lynne Rienner.

Baumol, William J. 1967. “Macroeconomics of Unbalanced Growth: the Anatomy of Urban

Crisis.” American Economic Review 57 (3): 415–426.

Bonelli, Regis, Samuel Pessoa, and Silvi Matos. 2013. “Desindustrialização no Brasil:
Fatos e interpretação.” In O futuro da indústria no Brasil, edited by Edmar Bacha and
Monica Baumgarten de Bolle, 45–79. Rio de Janeiro: Civilização Brasileira.

da Cruz, Marcio José Vargas, Gabriel Porcile, Luciano Nakabashi, and Fábio Dória
Scatolin. 2008. “Structural Change and the Service Sector in Brazil.” Working Paper No.
95. Universidade Federal do Paraná, Departamento de Economia.

De Negri, Fernanda, and Luiz Ricardo Cavalcante, eds. 2014. Produtividade no

(p. 104)

Brasil: Desempenho e determinantes. Brasília: IPEA and ABDI.

Dean, Warren. 1969. The Industrialization of São Paulo, 1880–1945. Austin: University of
Texas Press.

Furtado, Celso. 1972. Formação econômica do Brasil, 11th edition. São Paulo: Companhia
Editora Nacional.

Kaldor, Nicholas. 1966. Causes of the Slow Rate of Economic Growth of the United
Kingdom. Cambridge: Cambridge University Press.

McMillan, Margaret S., and Dani Rodrik. 2011. “Globalization, Structural Change and
Productivity Growth.” National Bureau of Economic Research, No. w17143.

Roett, Riordan. 1984. Brazil: Politics in a Patrimonial Society, 3rd edition. New York:
Praeger.

Page 20 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

Rowthorn, Robert, and Ramana Ramaswamy. 1999. “Growth, Trade, and

Deindustrialization.” IMF Staff Papers 46 (1): 18–41.

Silber, Simão David. 2011. “Foreign Trade and Foreign Investments: The Brazilian
Experience in the Last Two Decades.” In The Economies of Argentina and Brazil: A
Comparative Perspective, edited by Werner Baer and David Fleischer, 441–467.
Cheltenham, UK: Edward Elgar.

Skidmore, Thomas E. 1967. Politics in Brazil, 1930–64: An Experiment in Democracy.

New York: Oxford University Press.

Villela, André. 2011. “A Bird’s Eye View of Brazilian Industrialization.” In The Economies
of Argentina and Brazil: A Comparative Perspective, edited by Werner Baer and David
Fleischer, 38–65. Cheltenham, UK: Edward Elgar.

Notes:

(1.) The link of Brazil’s coffee support program to the import substitution of the 1930s
was first analyzed by Celso Furtado in his classic work Formação econômica do Brasil
(11th edition, 1972).

(2.) SUMOC was the forerunner of Brazil’s Central Bank.

(3.) Which would later add the word “social,” becoming the current BNDES.

(4.) Imports of capital goods as a percentage of total supply declined from 59% in 1949 to
12.9% in 1962; for intermediate goods it fell from 25.9% to 8.9%; and for consumer goods
it fell from 10% to 1.1%.

(5.) Critics of ISI pointed out that this dilemma was the result of across-the-board ISI,
without any consideration of being more selective by choosing only sectors with a
potential comparative advantage. The problem, however, was that many advanced
industrial countries were protective of their older industries, such as textiles and shoes.

(6.) For details on the political situation of the period, see Skidmore (1967, Chapter 6)
and Roett (1984).

(7.) For individual industry studies, see Baer (1969) and Amann (2000).

(8.) Some parts of this section are based on the article “Industrialization and De-
Industrialization in Emerging Economies: The Cases of Brazil and India” by Werner Baer
and Rahul A. Sirohi (2016).

Werner Baer

Page 21 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019

Brazil’s Import-Substitution Industrialization

Werner Baer, Professor of Economics, Department of Economics, University of Illinois

at Urbana-Champaign

Page 22 of 22

PRINTED FROM OXFORD HANDBOOKS ONLINE ([Link]). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Freie Universitaet Berlin; date: 14 January 2019