Bulmer-Thomas V., Coatsworth J., Cortes-Conde R. The Cambridge Economic History of Latin America: Volume 2, The Long Twentieth Century
Подождите немного. Документ загружается.
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c12BCB929-Bulmer 0 521 81290 9 September 30, 2005 15:41
532 Otto T. Solbrig
Here, plantations were established. Although it was not known at the time,
rubber trees in Amazonia suffer from the attack of a fungus. Only isolated
trees that escape the fungus survive. Because the fungus is not transmitted
through the seed, plantations in the Far East were disease-free. Plantations
allowed a much more rational and efficient handling of the trees and of
labor and allowed rubber to be produced at a fraction of the cost of tapping
wild trees in the Amazon. When plantations came into full production
shortly before World War I, the price of wild rubber collapsed and rubber
tapping in the Amazon became a marginal operation. Rubber, which had
been second only to coffee as an export crop in Brazil, suddenly became a
marginal product.
After World War II, Brazil moved its capital to the interior and embarked
on a concerted push to develop its interior. The population was growing and
moving in large numbers to the cities of the south, especially S
˜
ao Paulo.
Government policies were promoted to encourage agricultural develop-
ment. The modernization of agriculture in the southern states of S
˜
ao Paulo,
Paran
´
a, Santa Catarina, and Rio Grande do Sul forced many smallholders
to sell their land and move to the interior in search of new land. In this, they
were encouraged by an extensive program of colonization of the interior of
Brazil promoted by the government. The Amazon region, which includes
the savanna regions known as the cerrado, and the Amazon forest proper,
were targeted. Between 1970 and 1990, millions of new farmers settled in
these regions. EMBRAPA (Empresa Brasileira de Pesquisa Agropecuaria), the
federal agricultural research organization, developed a number of successful
new technologies for these regions. Much of Mato Grosso and southern
Goias were turned over to the growing of soybeans in large estates using
the most modern machinery and production methods. Smaller farmers in
frontier regions grew beans, rice, maize, and manioc. However, the greatest
area was dedicated to the creation of pastures and the raising of cattle. There
are now more cows than people in Brazil, which now has by far the largest
cattle herd in the world. New roads allowed the extraction of valuable trop-
ical timber. Large swathes of forest and savannas were transformed into
pastures or agricultural land, especially in the states of Par
´
a, Goias, Mato
Grosso, and Rond
ˆ
onia.
Similar advances on the Amazon forest took place in Peru, Colombia,
Bolivia, and Paraguay. The fastest rate of advance (but not the absolute
amount) of the agricultural frontier has been in Paraguay and Peru rather
than in Brazil. It is calculated that in another fifty years, most of the Amazon
forest will have been converted to other uses. Efforts are underway to protect
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c12BCB929-Bulmer 0 521 81290 9 September 30, 2005 15:41
Structure, Performance, and Policy in Agriculture 533
some of the enormous diversity of plants and animals by creating natural
reserves.
Among the new crops cultivated in these regions is the coca plant
(Erythroxylon coca) that, although illegal, is adapted to this environment
and is by far the most profitable crop. This is particularly so in Bolivia,
Peru, and Colombia, and to a lesser extent in Ecuador.
CONCLUSIONS
Over the last century, the principal change that can be observed is the
intensification of agriculture. Yet profitability per hectare is down, mainly
because of the increased cost of inputs and the relative fall in international
prices of most commodities.
The principal production costs of extensive agriculture practiced at the
beginning of the twentieth century were labor and land amortization. Seeds
were carried from one season to the next; fertilizer – if any was used –
was manure, also produced on the premises. Machinery was limited –
plows, seed drills, and harvesting equipment – and often produced locally.
Pesticides and herbicides were unknown. Intensive agriculture, on the other
hand, requires much purchased input. Seeds have to be bought, if not yearly,
at very frequent intervals. The new varieties require a better land preparation
that implies specialized machines. Tractors have replaced horses and oxen
as sources of power and, consequently, instead of fodder produced in the
premises, fuel has to be purchased. The new varieties also require the use
of chemical fertilizers, herbicides, and pesticides. Although the number of
agricultural workers is down, labor costs have come down much less. This is
in part because today’s worker has to be better prepared and in part because
of new labor laws, but also because of the need of technical advice that is
not cheap. Consequently, the farm’s net income is only 10 to 20 percent of
gross income. This makes producers that farm small areas noncompetitive
with large producers who can reap benefits of scale in the use of machinery
and in their capacity to obtain better prices for their inputs.
The dependence of countries on a few export commodities has not
changed throughout the long twentieth century, although the commodity
mix has. Coffee is still a major crop in Brazil, Colombia, and Central
America; sugar continues to be an important export product in Cuba and
Peru; cattle and cereals are very important in Argentina. To this list must
now be added soybeans and citrus in Argentina and Brazil; fruits and
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c12BCB929-Bulmer 0 521 81290 9 September 30, 2005 15:41
534 Otto T. Solbrig
vegetables in Chile and Argentina; and flowers in Colombia and Ecuador.
Still, prices of agricultural commodities have continued their downward
trend, especially for cereals.
Modern industrial agriculture is stacked against the small producer and
the producer of commodities, especially grains, but also coffee and sugar,
two high-value products. Synthetic products have also conspired against
the profitability of farmers. Synthetic rubber has replaced natural rubber;
synthetic fibers for cordage and twine have almost completely replaced sisal
and henequen; synthetic textile fibers have reduced significantly the market
for cotton and wool.
Another factor that reduces the profitability of agriculture is government
intervention. In developed countries, especially the United States, Western
Europe, and Japan, governments try to protect their farmers from compe-
tition from lower-cost producers. This takes many forms. In Europe, high
import taxes, export subsidies, and subsidies to farmers to produce cer-
tain commodities; in the United States, minimum prices are guaranteed; in
Japan, high tariffs keep imported products out. Farmers are sometimes paid
for not producing certain commodities or for producing them in certain
environmentally friendly ways. Export subsidies are also used generously.
In Latin America, especially in countries with large agricultural exports
such as Argentina, exports are taxed, thereby reducing the farmer’s profits.
Behind the complex web of subsidies in developed countries is the under-
standable wish of all societies for food security. Depending on imports for
food supply can make countries very vulnerable to external pressure. In
Latin American countries, there is a long tradition of relying on customs
duties as a major source of government revenue. Another tradition in Latin
America is the avoidance of taxes, a situation that almost forces governments
to use custom duties as a major source of revenue.
Subsidies and taxes distort markets and prices. In developing countries,
subsidies increase production and lead to the farming of marginal land.
The increased production lowers prices. Export taxes are equivalent to
increasing production costs. The reaction of Latin American farmers is
to produce more in compensation. This only makes the situation worse,
leading to overproduction.
The coffee agreements showed that it is impossible to regulate production
of commodities in a market with many actors and no dominant one to
impose market discipline. Overproduction and low prices are likely to
continue. From a Latin American point of view, the elimination of market
interventions – be they subsidies or taxes – would help create a fairer
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c12BCB929-Bulmer 0 521 81290 9 September 30, 2005 15:41
Structure, Performance, and Policy in Agriculture 535
situation. It also would be environmentally beneficial. This is not likely to
happen soon.
Another effect of low prices caused by market interventions by govern-
ments is to eliminate small producers. As margins decrease, small producers
with high fixed costs cannot compete and must find employment in larger
farms or outside of agriculture. This is part of the reason that after witness-
ing a decrease in average farm size during most of the twentieth century,
this trend has been reversed in the last thirty years.
What can we learn from the development of agriculture in Latin America
during the long twentieth century and what does it tell us regarding the
future? Agricultural and livestock products, together with mining, have
been the principal exports of Latin America since the arrival of the Spaniards
and Portuguese. During the first third of the long twentieth century, these
export products were the engine that pulled the development train, and it
led to rapid growth and, in some cases, most notably in the Southern Cone,
created a high standard of living. Yet, although it created pockets of wealth,
this phase was not sustainable. Until World War I, European countries and
the United States absorbed increasing amounts of cereals and beef, sugar
and coffee, rubber and henequen. But World War I brought this period to
an end and the world depression of the 1930s dealt it a mortal blow. Latin
America, following the analysis of the terms of trade by Prebischand led now
by anew more urban leadership, abandoned the export-led development
model in favor of import-substitution industrialization. Although in its
early phases the new model proved successful, it too was not sustainable
and eventually was abandoned in the 1980s.
Characteristics of Latin American agriculture during the long twentieth
century include change and diversity. From a traditional system based on
the large estate using extensive systems based on land and labor, agriculture
in the region has become much more capitalized, with a strong knowledge
base. In spite of several attempts at land reform, the large estate still dom-
inates the rural scene. Yet today’s estate differs considerably from that of
100 years ago. Gone are the semifeudal labor relations that prevailed at the
beginning of the twentieth century, replaced by a mechanized agriculture
that uses much less labor. The minifundio still exists in many regions, partic-
ularly in the Andean countries of Peru, Bolivia, and Ecuador, increasingly
pushed into a marginal situation both physically and socially.
Diversity was and continues to be a characteristic of the agriculture of
this vast region, a result of the variety of climates, topography, history, and
societies. So, not all large estates at the beginning of the twentieth century
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c12BCB929-Bulmer 0 521 81290 9 September 30, 2005 15:41
536 Otto T. Solbrig
conformed to the stereotype of the traditional hacienda, nor are all large
estates highly capitalized and efficient at the end of the period. The duality
between a modern export sector and a less developed domestic sector varied
considerably among countries and, in many cases, such as in Argentina,
did not exist.
The population of the world is still growing and so is its standard of living.
Future demand for agricultural products will therefore continue. The big
question is whether demand will grow faster than supply. Farmers, helped
by scientific research and new technologies, have been able to increase food
and fiber production at a rate faster than demand. Whether this kind of
agriculture is sustainable is debatable. And in the future, markets are likely
to demand sustainable production methods, presenting Latin America with
one more challenge.
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c13 CB929-Bulmer 052181290 9 October 6, 2005 10:44
13
THE POLITICAL ECONOMY
OF INDUSTRIALIZATION
stephen haber
This chapter builds on recent research by scholars across a broad number
of countries and disciplines to offer a reinterpretation of the history of
Latin American industrialization. It departs from the standard view that
underlies much of the literature on Latin America’s economic history: that
Latin America had open, “export economies” from the 1870stothe1930s,
and that these export economies were then dramatically transformed into
“import-substituting” economies during the period 1930–80.
Obviously, the history of industrialization was not the same in Bolivia
as it was in Brazil – there was considerable variation from country to coun-
try, with the smaller economies of the region generally industrializing later
and less completely than the larger economies. Capturing both the general
trends and the specifics of each case would exceed the space limitations of a
single chapter. I therefore concentrate my analysis on four cases: Argentina,
Brazil, Chile, and Mexico. I have chosen these cases because they account
for three quarters of total Latin American gross domestic product (GDP)
and because they are (and historically were) the most heavily industrialized
countries in the region. Other countries certainly have (and had) man-
ufacturing sectors but, as a practical matter, when we speak about Latin
American industry, we are really speaking about the manufacturing sectors
of Argentina, Brazil, Chile, and Mexico.
The reinterpretation that I offer contains four arguments. First, I argue
that there was substantial industry in Latin America well before 1930.In
fact, in the countries that we focus on in this chapter, the development
The author acknowledges helpful discussions with Herbert Klein, James Robinson, Kenneth Sokoloff,
and Barry Weingast, whose insights contributed to this chapter.
537
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c13 CB929-Bulmer 052181290 9 October 6, 2005 10:44
538 Stephen Haber
of large-scale, mechanized industry can be dated from the 1890s. Second,
one logical implication of this argument is that there must have been poli-
cies favorable to industrialization – particularly protective tariffs – during
the so-called era of export-led growth.
1
Third, Latin American govern-
ments, both before and after 1930,were protectionist not because of vague
economic goals, but because there were political benefits to supporting
manufacturers. The research by Ra
´
ul Prebisch and the United Nations Eco-
nomic Commission for Latin America (ECLA) economists in the 1950s gave
intellectual legitimacy to developments already taking place: their research
did not, in and of itself, cause governments to adopt policies designed
to protect and subsidize manufacturing. Fourth, the ultimate outcome of
import-substituting industrialization (ISI) is as depicted in the standard
literature: highly protected and woefully inefficient industries. The differ-
ence in the interpretation advanced here is that this outcome was not the
product of the rise of “developmentalist” states. It was the product of a
historically longer political process that involved economic agents making
specific demands on politicians who were not accountable to broad con-
stituencies. In particular, politicians were not accountable to the group that
ultimately paid the cost of protectionism: consumers, who tended to be
politically unorganized and who typically lacked electoral mechanisms that
would allow them to sanction public officials for policies inimical to their
interests.
This is not to argue that there were no differences between the pre- and
post-1930 periods of Latin American industrial development. First, man-
ufacturing was quantitatively larger in the post-1930 period. Second, in
the pre-1930 period, virtually all industry was domestically owned. In the
post-1930 period (and especially after 1950), domestic firms were joined
by multinational enterprises. Third, after 1930, industrialists had a pow-
erful political ally in organized labor. Industrialists and labor unions may
not have liked one another (indeed, they actively abhorred one another),
but they did have a common interest in trade protection. They therefore
formed coalitions that would only be broken with the collapse of import
substitution in the 1980s. Even noting these differences, however, it is clear
that there is no neat divide between the pre- and post-1930 periods in terms
1
One could dispense with this logical implication by arguing that Latin American countries had a
comparative advantage in manufacturing in the early twentieth century. Such an argument would,
however, be difficult to sustain. Indeed, the evidence all points the other way. See, for example,
Gregory Clark, “Why Isn’t the Whole World Developed? Lessons from the Cotton Mills,” Journal of
Economic History 47, 1 (1987): 141–74.
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c13 CB929-Bulmer 052181290 9 October 6, 2005 10:44
The Political Economy of Industrialization 539
of government policies, the adoption of mechanized technologies, the scale
of enterprise, or the inward-looking nature of industry.
Pushing back the point at which governments initiated pro-industry
policies creates somewhat of a conundrum. Latin American governments
were supposedly great supporters of laissez faire in the late nineteenth and
early twentieth centuries. How, then, could governments simultaneously
embrace laissez faire and protectionism? The answer is that governments
do not have to frame coherent policies. The notion that governments are
“autonomous” entities that frame development “strategies” is social-science
fiction. A more realistic view is that self-interested economic agents make
endless demands on public officials to create policies that confer an eco-
nomic benefit on them and them alone. When granting those policies
is in the self-interest of public officials, they do so. The accretion of ad
hoc policies means that the overall package of regulations, laws, and tax
codes that comprise a property rights system may work at cross-purposes to
one another. The policies that supported industrial development in Latin
America – both during the so-called periods of “export-led growth” and
“import-substituting industrialization” – were not an exception to this gen-
eral rule.
INDUSTRIAL DEVELOPMENT AMIDST
EXPORT-LED GROWTH, 1890–1930
There had been, at least since the mid-nineteenth century, precocious
attempts at industrial development in a number of Latin American coun-
tries. Nevertheless, the available quantitative evidence indicates that what-
ever industry had been founded grew at a very modest pace. Per capita
incomes were too low, and markets were too isolated by high transport
costs, to support modern manufacturing.
2
The pace of industrial growth accelerated rapidly in the 1890s. Circa 1914,
the larger countries of the region – particularly Brazil, Mexico, Chile, and
2
See, for example, Stanley Stein, The Brazilian Cotton Manufacture: Textile Enterprise in an Underde-
veloped Area, 1850–1950 (Cambridge, MA, 1957); Stephen Haber, “Assessing the Obstacles to Industrial
Development, The Mexican Economy,” Journal of Latin American Studies 24, 1 (1992): 1–32; Armando
Razo and Stephen Haber, “The Rate of Growth of Productivity in Mexico, 1850–1933:Evidence from
the Cotton Textile Industry,” Journal of Latin American Studies 30, 3 (1998): 481–517;Fernando
Rocchi, “Building a Nation, Building a Market: Industrial Growth and the Domestic Economy in
Turn-of-the-Century Argentina” (Ph.D. dissertation, University of California, Santa Barbara, CA,
1997).
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c13 CB929-Bulmer 052181290 9 October 6, 2005 10:44
540 Stephen Haber
Argentina – boasted sizable manufacturing sectors, which produced a broad
range of consumer nondurables (particularly beer, cigarettes, soap, matches,
hats, paper, footwear, and cotton cloth). In some cases, domestic firms
had also moved into the production of intermediate inputs (basic chemi-
cals, glass bottles) and construction goods (particularly cement, explosives,
bricks, steel rails, and iron and steel structural shapes).
The impetus for industrial development came from the expansion of
foreign trade. Driving the growth of foreign trade were two factors. The
first was that most Latin American countries were on the silver standard,
and silver fell in value relative to gold in the last two decades of the nine-
teenth century. Most Latin American countries, therefore, saw their cur-
rencies depreciate in real terms relative to the gold-backed currencies of
the economies of the North Atlantic. As international trade theory would
predict, real exchange rate depreciation resulted in the expansion of the
tradables sectors at the expense of nontradables. Second, the late nine-
teenth century also saw a dramatic decline in the international costs of
transport, as steel-hulled steamships came to replace wood and sail.
Real exchange rate depreciation and falling costs of maritime transport
kick-started Latin American economic growth. Latin American countries
could now produce and ship goods at a price low enough to be competitive
in the United States and Western Europe. These exports included min-
erals (particularly copper and lead), industrial fibers (cotton, wool, jute,
and sisal), staple agricultural goods (primarily beef and wheat), and non-
staple agricultural commodities (the most important of which were coffee,
bananas, and sugar). The result was a wave of foreign direct investment in
precisely those sectors. In order for these investments to bear fruit, how-
ever, there had to be a means to move commodities from their point of
production to the ports or, in the case of Mexico, to the U.S. border. Trade
therefore gave rise to the construction of railroad networks (also financed
by foreign investment) that crisscrossed Latin America, linking major cities
and integrating markets.
3
3
Most of the best work done to date on railroads focuses on the cases of Brazil and Mexico. See, for
example, the chapter in this volume by William Summerhill. Also see his “Transport Improvements
and Economic Growth in Brazil and Mexico,” in Stephen Haber, ed., How Latin America Fell Behind:
Essays on the Economic Histories of Brazil and Mexico, 1800–1914 (Stanford,CA, 1997), 93–117, and Order
Against Progress: Government, Foreign Investment, and Railroads in Brazil, 1854–1913 (Stanford, CA,
2003). On Mexico, see Sandra Kuntz Ficker, Empresa extranjera y mercado interno: el Ferrocarril Central
Mexicano, 1880–1907 (Mexico City, 1995); Sandra Kuntz Ficker and Paolo Riguzzi, eds., Ferrocarriles y
vida econ
´
omica en M
´
exico, 1850–1950:Del surgimiento tard
´
ıo al decaimiento precoz (Mexico City, 1996);
and John H. Coatsworth, Growth Against Development: The Economic Impact of Railroads in Porfirian
Mexico (Dekalb, IL, 1981).
Cambridge Histories Online © Cambridge University Press, 2008
P1: GDZ
0521812909c13 CB929-Bulmer 052181290 9 October 6, 2005 10:44
The Political Economy of Industrialization 541
The growth of the export sector, the concomitant growth of railway net-
works, and implicit trade protection created by depreciating real exchange
rates created the conditions conducive to industrial development. First,
there came into existence a wage-earning population of ranch hands, min-
ers, plantation workers, stevedores, and railway men that generated con-
siderable demand for consumer goods. Second, this growing population
of working-class consumers was not made up of isolated pockets of work-
ers toiling away in enclaves: they were part of a national market that was
linked by railways. In fact, as William Summerhill has shown for Brazil
and Sandra Kuntz Ficker has shown for Mexico, most of the commodities
carried by Latin American railways were destined for domestic (not interna-
tional) markets.
4
Finally, depreciating real exchange rates operated exactly
the same as an across-the-board increase in import tariffs: they raised the
domestic price of imported goods.
Latin America’s merchants, who earlier had been importing manufac-
tured goods, quickly realized the potential created by growing markets and
implicit trade protection. They soon began to erect textile mills, beer brew-
eries, cigarette factories, soap works, and other manufacturing enterprises.
The production of consumer goods, in turn, produced demand for inter-
mediate inputs, such as glass bottles, paper, and basic chemicals. At the
same time, the growth of ports and railways meant that there was demand
for construction goods, such as steel and cement. Almost immediately,
Latin America’s nascent industrialists began to pressure their governments
for high tariffs on the final goods they produced (in order to further protect
them from foreign competition) and for low or zero tariffs on the capital
and intermediate goods they needed in order to ramp up production.
This process of industrial development as a consequence of the growth
of the export sector has been most intensively studied in the case of Mexico.
Prior to the 1880sMexican manufacturing was hampered by low per capita
incomes and by the lack of long-distance transport. Beginning in the 1880s
and accelerating in the 1890s, “export-led growth” removed these obsta-
cles to industrial development. As a consequence, industry expanded at a
dramatic pace.
5
4
William Summerhill, “Transport Improvements and Economic Growth in Brazil and Mexico,” and
Order Against Progress, ch. 6;Sandra Kuntz Ficker, Empresa extranjera y mercado interno.
5
Foradetailed discussion, see Stephen Haber, Industry and Underdevelopment: The Industrialization
of Mexico, 1890–1940 (Stanford, CA, 1989), and “Assessing the Obstacles to Industrial Development;”
Mario Cerutti, Burgues
´
ıa, capitales, e industria en el norte de M
´
exico (Monterrey, Mexico, 1992); Dawn
Keremetsis, La industria textil mexicana en el siglo XIX (Mexico City, 1973); Carlos Marichal and
Mario Cerutti, eds., Historia de las grandes empresas en M
´
exico, 1850–1930 (Mexico City, 1997); Aurora