The Mechanics of Brazil’s Auto Industry

In April, 1994, the vice-president of General Motors
of Brazil (GMB) made a special appearance on
prime-time television. Instead of the standard sales
pitch, he asked viewers to stop buying the Corsa, GM’s
hottest-selling car. GMB was forced to make this
unprecedented appeal by the overwhelming response to
its new compact. Only two months after appearing on
the market, waiting lists for the Corsa already
approached 130,000, nearly all of planned production
for 1994. If Brazilians did not wait until local produc-
tion caught up to demand, GM feared that the illegal
30-50% markups being charged by disreputable dealers
and in the burgeoning black market would rise even
further.
After more than a decade of stagnation, Brazil’s auto
market has boomed. Total vehicle production grew by
30% in 1993 and another 14% in 1994, making 1994
the most profitable year in GMB’s 69-year history.
Optimistic about the future, GM Brazil is planning to
invest US$2 billion over a three-year period, all from
locally generated profits. Brazil has moved to the fore-
front of GM’s global strategy, a far cry from when
GM’s Chief Executive Officer from the 1980s, Roger
Smith, disparagingly referred to Brazil as a “black
hole.”
GM is not alone in its renewed enthusiasm for Brazil.
Ford, Fiat and Volkswagen, which, together with GM,
dominate the local industry, are also planning to invest
US$9-12 billion in modernization and new capacity.
They will be joined by Renault, Hyundai and Asia
Motors, the first new entrants into the car market since
the early 1970s. Mercedes, which previously produced
trucks and buses, will start assembling passenger cars.
Toyota has expressed similar interest in entering
Brazil’s passenger-car market. While Brazil was the
world’s ninth-largest producer in 1994 with an annual
output of 1,581,381 vehicles [see table, p. 33], some
predict that by the year 2000, it will produce two mil-
lion cars. This would move Brazil into sixth place, sur-
passing England, Canada and South Korea.
Helen Shapiro is an economist who teaches at Harvard
University’s Graduate School of Business Administration. She is
the author of Engines of Growth: The State and Transnational
Auto Companies in Brazil (Cambridge University Press, 1994).
Neoliberal trends aside, the Brazilian
state has developed specific sectoral
policies to revitalize the auto industry.
These policies-not market
liberalization-account for the
industry’s recent boom.
The recent recovery of Brazil’s auto industry coin-
cides with the reduction of trade barriers and deregula-
tion, leading many to conclude that market liberaliza-
tion is driving the boom. This conclusion is unwarrant-
ed, however. While selective reforms have been criti-
cal, the revival of the auto industry is not the result of
free trade and a noninterventionist state. In fact, despite
the overall liberalizing trend since 1990, the Brazilian
state has developed specific sectoral policies aimed at
promoting the industry’s recovery, including organized
negotiations between the government, the auto compa-
nies and labor. It is these deliberate state policies in
concert with a changing industry and international eco-
nomic context that explain the auto industry’s recent
boom. Indeed, Brazil’s auto industry has been largely
shaped by the dynamic interaction between state policy
and the global auto industry since its beginnings in the
1950s.
The auto industry has many of the attributes of a clas-
sic oligopoly. It is capital intensive with large
economies of scale. To be cost-effective, an integrated
auto plant must produce about 250,000 units a year. Its
requirements of large fixed investments, distribution
and service networks, and brand-name recognition cre-
ate enormous barriers to entry for newcomers. As a
result, it is dominated by a handful of transnational cor-
porations whose strategies are interdependent. A firm’s
decision about where to produce, for example, may
have more to do with its market position relative to a
competitor than with simple short-term profit or cost
calculations. I
Despite images conjured up by the development of
the “world car,” in which models are standardized for
world markets and car components come from different
NACLA REPORT ON THE AMERICAS 28REPORT ON TRANSNATIONAL INVESTMENT
An advertisement for the Corsa,
General Motors of Brazil’s hottest-
selling car.
locations around the world, auto companies do not ran-
domly survey the globe in search of low-wage produc-
tion sites. This may be an accurate portrayal of firm
strategy in labor-intensive industries such as textiles
that can easily relocate to exploit low wages and gov-
ernment incentives, and where low-entry barriers allow
new competitors to challenge firms operating from
high-cost locations. By contrast, capital-intensive
industries like auto depend on an industrial infrastruc-
ture for parts and components, skilled labor, and a siz-
able domestic market. Wage rates, while important, are
only part of the story.
In Latin America, auto production is concentrated in
the larger, more industrially advanced countries of
Brazil, Mexico and Argentina. Though tiny in terms of
global market share-3%, 2%, and 1%, respectively–
the auto industry dominates manufacturing in each
country’s economy. In the past five years, these Latin
American markets all experienced booms in demand,
in contrast to almost nil growth in the developed coun-
tries. Given the low rates of car ownership– to every
11.6 people in Brazil, 1 to 8.5 in Mexico, and 1 to 5.5
in Argentina, as compared to 1 to 3 in the United
States-transnational auto companies were attracted to
the huge growth potential of these overseas domestic
markets.
he roots of the modern Brazilian auto industry
can be traced back to 1956, when the govern-
ment of Juscelino Kubitschek (1956-1961)
implemented a five-year plan initially motivated by
balance-of-payments concerns and the desire to protect
local-parts producers that had emerged during the
Second World War. The Kubitschek administration
came to see the auto industry as
the quickest, most effective way
of promoting Brazil’s industrial-
ization: “fifty years (of develop-
ment) in five” was his campaign
slogan. By attracting foreign cap-
ital and technology and generat-
ing linkages to complementary
sectors, the auto industry was to
be the leading sector in a broad
import-substituting industrializa-
tion push.
The plan’s basic approach was
to restrict imports and force
transnational automotive compa-
nies to choose between abandon-
ing the Brazilian market or pro-
ducing vehicles with 90-95% Brazilian-made content
within five years. Until that point, Brazil’s auto indus-
try consisted of foreign subsidiaries or licensed domes-
tic firms that assembled vehicles locally from imported
completely or semi-knocked down kits. Transnational
corporations expressed little interest in shifting from
assembly to full manufacture. As a Ford representative
put it at the time, the idea of producing engines in the
“tropics” was “utopian.”
But a combination of factors made full-scale car
manufacturing in Brazil more feasible. At the interna-
tional level, as the postwar boom came to an close,
profits on domestic operations fell in both the United
States and Europe, pushing firms to place greater
emphasis on overseas expansion. International compe-
tition intensified as a result, and aggressive European
firms began to challenge the U.S. Big Three (GM, Ford
and Chrysler) in their traditional markets including
Latin America. When one firm took the first step by
deciding to invest in Brazil, other firms followed in
order to defend their potential market shares. 2
The dynamics of oligopolistic competition were not
the only factor, however. The resulting shake-up in the
global auto industry gave the Brazilian government
greater leverage vis-i-vis the auto transnationals. By
issuing a credible threat that the Brazilian market
would be closed to imports, the government was able
to pressure the transnationals to engage in full-scale
local production. The government also offered exten-
sive subsidies that significantly reduced the cost of
capital investment and guaranteed a return even if prof-
its did not materialize. Since most of these financial
incentives were available for only the five years of the
plan, the government set the timing of firm investment
Vol XXIX, No 4 JAN/FEB 199629 Vol XXIX, No 4 JAN/FEB 1996 29REPORT ON TRANSNATIONAL INVESTMENT
by increasing the barriers to entry for firms that wanted
to delay investing. By compelling large investments,
the government increased the exit costs as well. Eleven
firms signed up. Three (Willys-Overland, Vemag and
the National Motor Factory) were controlled by
Brazilian capital, and two (Mercedes-Benz and Simca)
were joint ventures. The rest were controlled by or were
wholly owned subsidiaries of foreign firms.
The early 1960s marked a period of economic insta-
bility and labor unrest in Brazil, culminating in the mil-
itary coup of 1964. For the auto industry, this was a
period of crisis. Firms had built ahead of demand, and
the industry became plagued with overcapacity. The
auto sector was further hard hit as the market shrank
with the 1963-64 economic downturn and the military’s
post-coup austerity program. Smaller, financially weak-
er firms did not survive these lean years. By 1968, the
original eleven firms had shrunk to eight. Only those
controlled by transnational capital remained, of which
three-Volkswagen, GM and Ford-controlled 89% of
the total vehicle market.
Workers were brutally repressed during the military
regime, and ultimately forced to shoulder the costs of its
austerity program. Workers’ real wages fell drastically
during this period, and their share of total income was
redistributed to the top income brackets. This income
concentration, along with the military’s easing of tight
credit policies after 1967, paved the way for an econom-
ic boom based on consumer durables such as automo-
biles and household appliances. Brazil’s GDP grew 10%
annually between 1968 and 1974. The auto industry was
key to jumpstarting this so-called “economic miracle” by
sustaining 20% annual growth rates between 1968 and
1973. Excess capacity, once a liability, now allowed
firms to meet rising demand. By the early 1970s, pro-
duction capacity couldn’t keep pace with demand, lead-
ing firms to initiate a new wave of investments.
Changing international conditions made it diffi-
cult to maintain the high “miracle” growth rates.
The oil shock in 1973 came down heavily on
Brazil, which imported 80% of its oil needs. The con-
servative military government, concerned about the
growing trade deficit, sought-in addition to heavy for-
eign borrowing-to move the auto industry towards
exports as a way of obtaining needed foreign exchange.
For the transnational subsidiaries, which had invested
heavily in the domestic market with the expectation of
continued rapid growth, exports served as a temporary
outlet for excess capacity.
Once again, changes in the global industry reinforced
the government’s objectives, this time toward exports.
Powerful new competitors from Japan, whose revolu-
tionary production technique produced low-cost, high-
quality vehicles for export, challenged the existing
Auto workers on the first assembly line in So Bernardo do
Campo, Sao Paulo in 1956.
international market structure. GM and Ford, for exam-
ple, sought to catch up by automating at home. They
also adopted “world car” strategies, which allowed
them to produce at higher volumes and lower cost by
increasing product standardization worldwide. 3 Such
global production networks allowed the companies to
increase economies of scale, spread research and devel-
opment costs over more vehicles, and use low-wage
production sites as export platforms for engines and
components. Some European firms also looked to
developing countries, such as Spain and South Korea,
as low-cost export bases for finished vehicles for other
low-income countries with similar demand profiles.
The military government in Brazil promoted the
development of cars for export by launching the Special
Fiscal Benefits for Exports (BEFIEX) program in 1972.
The government granted generous incentives, including
tax exemptions on imported machinery, equipment and
other parts, and waived federal and state value-added
taxes on exports. In exchange, firms had to commit to
long-term export contracts and comply with minimum
domestic-content requirements (85% for vehicles sold
in Brazil). 4 Firms were also allowed to import a certain
number of parts and components that had previously
been banned because they were produced domestically.
The Brazilian government was able to leverage priv-
ileged access to its large and growing domestic market
to pressure the auto transnationals to produce cars and
components for export in Brazil. For example, Fiat,
which until then had no presence in Brazil, was allowed
to enter the domestic car market only in exchange for
exporting 155,000 engines annually. In terms of auto
firms already in Brazil, the state had greater clout since,
in contrast to the 1950s, these firms had to protect not
only their access to the Brazilian market, but their prior
investments as well.
30NACLA REPORT ON THE AMERICAS NACIA REPORT ON THE AMERICAS 30REPORT ON TRANSNATIONAL INVESTMENT
For example, GMB want-
ed to introduce the J car,
one of GM’s “world cars,”
The Brazilian to Brazil. However, the
size of the domestic mar-
state in the ket did not justify the
1950s
saw the
investment
needed to build
the J car’s new engine,
auto industry as which had to be made in
Brazil in order to comply
the quickest, with domestic-content
requirements.
GM decided
to absorb the excess capac-
way of
ity while the domestic
market grew by exporting
promoting the the engine to the U.S.
Pontiac division. Exiting a
country’s market can be costly, as
industrialization, can reentry. The fact that
firms were forced to con-
sider Brazil as a produc-
tion site for export, when
they otherwise might not
have, speaks to a successful dynamic outcome of
import-substituting policies from the 1950s and early
1960s. Over the course of the 1970s, Brazil’s total auto-
motive exports had increased from virtually zero to
US$1 billion. 5
By 1980, the industry was churning out over a mil-
lion vehicles a year, making Brazil the world’s eighth-
largest producer. There was also a surge in trade-union
activity in the auto sector during this period, as work-
ers’ protests against declining wages and authoritarian-
ism in the workplace grew and eventually linked up to
broader opposition to the military government.
Although exports had grown to 14% of production,
firms continued to focus on domestic demand, which
was predicted to hit two million by 1990. With volume
up and new investment coming on stream, costs were
declining. Even the World Bank concluded that the
industry was a successful “infant” on the verge of matu- rity. 6
The debt crisis of the early 1980s snuffed out these
optimistic projections. Once again confronted with
domestic-market contraction and macroeconomic
uncertainty, the industry stagnated over the 1980s.
Firms looked to exports as an alternative, and by 1990, when the BEFIEX program was phased out, Brazil’s
car exports reached US$8.2 billion. 7 Exports came to
substitute for, rather than complement, domestic sales.
The direction of these exports began to shift as
Brazil’s Third World markets also contracted in
response to debt-driven austerity programs and
increased protectionism. In the early 1970s, nearly 90%
of Brazil’s finished-vehicle exports went to other Latin
American countries. By 1984, 40% of Brazil’s vehicle
exports were being shipped to the U.S. and Europe, and
by 1989, this share had increased to 60%. Brazil’s
biggest success stories were the Volkswagen Fox in
North America and the Fiat Duna in Europe. For a brief
time, it seemed that Brazil might be able to duplicate
South Korea’s strategy of exporting cheap cars to these
profitable markets. 8 However, the increasingly overval-
ued Brazilian currency pushed these cars out of their
price-sensitive market niche.
By decade’s end, Brazil produced fewer than a mil-
lion vehicles annually. New investment had ceased, and
many firms’ profits were increasingly derived from the
lucrative financial market rather than car sales. Within
the global industry, Brazil was eclipsed by Spain, South
Korea, and even Mexico, which outproduced Brazil for
the first time in 1991.
n the 1990s, an intensive restructuring process has
reshaped the global auto industry. As markets
become more open and brand-name loyalties weak-
en, firms are less likely to design unique car models for
particular national markets. Companies are integrating
their global operations in an attempt to eliminate dupli-
cation and cut costs. This has reshaped the transnation-
al auto manufacturers’ relations with their parts suppli-
ers, for example. Firms are now turning to outside sup-
pliers to do some of the processes that they once did in-
house. In imitation of the Japanese, they are also reduc-
ing the number of parts suppliers they work with in an
attempt to collaborate more closely on product design
and development. The pressure on these firms to meet
international standards with respect to cost, quality and
delivery time has increased. 9
Recent Brazilian governments have taken steps that
reinforce the industry trend of global integration. The
failure of the heterodox economic policies of President
Jos6 Sarney (1985-90) and subsequent bouts of hyper-
inflation paved the way for the implementation of
neoliberal economic policies in Brazil. President
Fernando Collor de Mello (1990-92) initiated neoliber-
al reforms, including his campaign promise to open the
Brazilian market to imported cars for the first time
since the late 1950s. Competition from imports was
meant to jumpstart the stagnating domestic auto indus-
try by forcing firms to invest in new technologies and
update locally produced models, which Collor said
resembled “horse-drawn buggies.” The Collor adminis-
tration reduced domestic-content requirements from
90% to about 70% and loosened other regulations pro-
tecting domestic suppliers in an effort to give firms
more flexibility with respect to sourcing and to reduce
the time required to introduce new models.
The removal of the ban on imports represented a sea
change in Brazilian policy. In order to compete against
Vol XXIX, No 4 JAN/FEB 199631 Vol XXIX, No 4 JAN/FEB 1996 31REPORT ON TRANSNATIONAL INVESTMENT
imports, the transnational subsidiaries in Brazil have
been forced to modernize their plants. They are moving
towards state-of-the-art technology, the automation of
more processes, and the introduction of new models in
an effort to improve both quality and
productivity. Longer production runs,
made possible by increasing sales,
have helped improve efficiency. They
have also reduced the number of main
suppliers and forced those remaining
to cut costs.
Nevertheless, the industry’s rebound
cannot be characterized as a simple
consequence of neoliberal reform.
Other state policies implemented over
the past five years that go against the
grain of the general liberalizing trend
have played a key role in reshaping
and revitalizing the Brazilian auto
industry in the 1990s. For example,
despite orthodox policy recipes to the
contrary, firms must still satisfy high
domestic-content requirements. Like-
wise, tariffs in the automotive sector
remain higher than in other industries An adrtisementruck.
and are being reduced only gradually. up truck
Far from pulling back, the Brazilian state continues to
actively manage trade through industry-specific trade
accords such as Mercosur, a customs union linking
Brazil, Argentina, Paraguay and Uruguay. That accord
facilitates the use of Brazil as an export platform for
cars for the region.
Perhaps the most important policy initiative designed
to promote the auto industry’s recovery was the cre-
ation of sectoral chambers that bring together the state,
the auto firms, and labor unions. The chambers were
convened by the Collor administration, which feared
that the negative impact of neoliberal economic
reforms on the industry would lead to plant shutdowns
similar to the one carried out by Ford in early 1992. In
the initial agreement, signed in 1992, in exchange for
government promises to reduce state and federal taxes,
the auto firms agreed to reduce costs and profit mar-
gins-and ultimately, prices-and to invest approxi-
mately US$20 billion by the year 2000. The firms also
agreed to increase employment and to guarantee a real
wage increase of 20% from 1993-95 in exchange for
labor’s implicit agreement to engage in collective bar-
gaining before going out on strike. The agreement led
to a 22% reduction in retail prices. I0
The chambers, which started as an emergency mea-
sure to save the auto industry from collapse, evolved
into an ongoing forum in which the state, business and
labor attempted to negotiate realistic policies that all
three sides could agree to. In 1993, the sectoral agree-
fo
ment was expanded to include a plan to develop “pop-
ular cars,” so named because their small size (engines
less than 1000 cc) and low price (around $7,200) made
them affordable to Brazil’s middle class. Itamar
Franco’s government (1992-1995)
agreed to tax reductions for these
small cars in exchange for commit-
ments from firms to invest and from
labor to moderate its demands.
The popular-car program shifted
demand toward small cars, to 47% of
overall car purchases, compared to
30% in 1993. Exports to Argentina
have skyrocketed to 69% of total
vehicle exports. Surprisingly, the
market recovered and firms planned
new investments in 1993, despite a
2,700% inflation rate. A five-percent
economic growth rate that year was
more compelling than the lack of
price stability or fiscal reform.
The boom in auto production has
not strengthened labor’s hand. As a
result of increased automation and Chevrolet’s pick- subcontracting-and lots of overtime
put in by auto workers on the factory
floor-the industry has been able to produce almost
80% more vehicles with 11% fewer workers compared
with 1990. ‘ Not only has employment not increased, as
the sectoral chamber promised, but wage increases
haven’t been forthcoming either. With the adoption of
tight credit policies by the new administration of
Fernando Henrique Cardoso, sales fell sharply after
mid-1995, further undermining labor’s bargaining posi-
tion. Some 150,000 workers were laid off in Sio
Paulo’s automotive industry between June and
November, 1995.12
The state has, on occasion, sidestepped the sectoral
chamber in the interest of larger macroeconomic con-
cerns, to the dissatisfaction of both the auto firms and
the trade unions. For example, in October, 1994, the
exiting Franco government unilaterally reduced the
import tariff on cars from 35% to 20% several months
earlier than planned. The move was motivated in part to
make it harder for auto transnationals to carry out their
promised wage increase. The government feared the
firms would pass the cost on to car buyers, thereby
refueling inflation. By the end of 1994, imports
accounted for 25% of car sales.
While the Cardoso administration continues to pro-
mote the sectoral chamber as a way of managing indus-
trial policy, it too has sacrificed negotiated settlements
to macroeconomic concerns. An increase in popular-car
taxes was motivated at least in part by the government’s
fear that it was losing needed revenue. In the shadow of
32 NACLA REPORT ON THE AMERICASREPORT ON TRANSNATIONAL INVESTMENT
the Mexican financial crisis in December, 1994, the
Brazilian government worried that the combination of
an appreciating currency and widening trade deficits
would lead investors to withdraw funds in anticipation
of a currency devaluation. Troubled by the “orgy” of
car imports, which was feeding growing monthly trade
deficits, Cardoso put tariffs on imports back up to 32%
in February, 1995. Imports continued to surge, howev-
er, leading the government to increase the tariff drasti-
cally to 70% in April. Some firms were hit harder than
others, particularly those that depended on imports to
supplement their domestic product line. The very fast
growth rates of early 1995, which fueled the auto indus-
try’s boom, threatened Cardoso’s stabilization program,
which had successfully reduced monthly inflation to
less than 2%. The Cardoso administration has since
engineered an economic slowdown by raising interest
rates, which has cut into auto sales.
overnment planners in the 1950s expected that
forging an auto industry made up of transna-
tional firms would make it easier for a country
like Brazil to access technology and foreign markets.
Due to political and economic conditions, financial
constraints, and the nature of
the global industry at the time, World Motor VI building a national industry (ToI with domestic firms was never
seriously considered. Domestic Country Tota parts producers expressed little United States interest in expanding into vehi-
cle production and the one Japan
state-owned firm was ill-suited Germany
to the task. Moreover, the U.S. France
Big Three were unwilling to Canada
license technology at that time. South Korea
The Brazilian strategy con- Spain
trasts to that of South Korea, United Kingdom which actively promoted Brazil domestically owned firms in Italy the 1970s. 1 3 While that ap- China
proach allows for more nation- ex-Soviet Union al control, it required huge investments and access to Mexico
licensed technology. The India
Korean government’s control Belgium
over foreign-exchange alloca- Sweden
tion and access to cheap financ- Taiwan
ing gave it more leverage over Poland domestic firms. Korean firms Australia also had access to foreign tech- Indonesia nology through licensing agree-
ments and strategic alliances Source: Compiled by the
with both Japanese, and later, Manufacturers Associatic
U.S. transnationals. The quick-
The Mechanics of Brazil’s Auto Industry
1. See David B. Yoffie, ed., Beyond Free Trade: Firms, Governments,
and Global Competition (Boston: Harvard Business School Press,
1993).
2. For more on this period and why the Brazilian government was
both unwilling and unable to limit the number of firms, see
Helen Shapiro, Engines of Growth: The State and Transnational
Auto Companies in Brazil (Cambridge: Cambridge University
Press, 1994).
3. For an early account of the “world car” and its implications for
Latin America, see “The World Car: Shifting into Overdrive,” in
NACLA Report on the Americas, Vol. 13, No. 4 (July/August,
1979).
4. N. Lee and J. Casson, “Automobile Commodity Chains in NICS:
A Comparison of South Korea, Mexico and Brazil,” in Gary
Gereffi and Miguel Korzeniewicz, eds., Commodity Chains and
Global Capitalism (Westport: Praeger, 1994).
5. Dani Rodrik, “Taking Trade Policy Seriously: Export Subsidization
as a Case Study in Policy Effectiveness,” NBER Working Paper
No. 4567, Cambridge, MA.
6. World Bank, Brazil: Industrial Policies and Manufactured Exports
(Washington, DC: World Bank, 1983).
7. Rodrik, “Taking Trade Policy Seriously.”
8. Analogy made in James P. Womack, Daniel T. Jones and Daniel
Roos, The Machine that Changed the World (New York: Rawson
Associates, 1990).
9. Anne Carolina Posthuma, “Restructuring and Changing Market
Conditions in the Brazilian Auto Components Industry,” ECLAC/
IDRC, Santiago, Chile, 1995. For background on the autoparts
industry, see Carin Addis, “Local Models: Auto Parts Firms and
Industrialization in Brazil,” Ph.D. dissertation, MIT, 1993.
10. The Economist Intelligence Unit, International Motor Business,
October, 1992.
11. Veja (Sao Paulo), November 1, 1995.
12. Veja (Sao Paulo), November 1, 1995.
13. For more on why the Brazilian government opted for this strategy
and how it contrasts with South Korea’s, see Shapiro, Engines of
Growth, and “Automobiles: From Import Substitution to Export
Promotion in Brazil and Mexico,” in Yoffie, Beyond Free Trade.