Jump to ContentJump to Main Navigation
The Economic Development of Latin America since Independence$

Luis Bértola and José Antonio Ocampo

Print publication date: 2012

Print ISBN-13: 9780199662135

Published to Oxford Scholarship Online: January 2013

DOI: 10.1093/acprof:oso/9780199662135.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 21 April 2018

Latin America in the World Economy, 1810–2010

Latin America in the World Economy, 1810–2010

Chapter:
(p.1) 1 Latin America in the World Economy, 1810–2010
Source:
The Economic Development of Latin America since Independence
Author(s):

Luis Bértola

José Antonio Ocampo

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199662135.003.0001

Abstract and Keywords

Chapter 1 presents an overview of Latin American development since Independence in a comparative perspective. Based on a new dataset of per capita GDP and a typology of Latin American countries, different stages of development and episodes of divergence and convergence with world leaders are identified. High volatility, linked to the pattern of production specialization and to pro-cyclical movements in capital flows, is identified as an important feature of Latin American development. Different stages of development are identified based on the forms of integration into the world economy. The environmental features of development are considered. Lags and achievements in schooling and life expectancy are discussed, as part of a broader analysis of a new historical human development index. The chapter ends with a discussion of the role played by inequality in Latin American development.

Keywords:   comparative development, convergence/divergence, economic volatility, human development, inequality, capital flows, balance of payments constraints, environment

Introduction

Anybody who has written about the economic history of Latin America inevitably starts out by wondering whether it is possible to make any generalizations at all about such a large continent, one that stretches so far north and south, with all the climatic differences which this entails, that is marked by vast mountain chains, deserts, and rainforests, and that displays such striking variety in terms of its geography, climate, and natural resources. It has also been the birthplace of a wide diversity of cultures which have, in turn, undergone radical changes wrought by colonization, emigration and immigration (both voluntary and involuntary), trade, and technology transfer.

Yet despite all of this, it does seem possible to talk about the economic history of Latin America as a whole and to discern a set of features which Latin American countries share, and these features can serve as a basis for arriving at an understanding of the direction in which their economies and societies have evolved. In this chapter we will attempt to describe some of what appear to be the most important characteristics of this type. We will also, however, try to refine these generalizations in two ways. On the one hand, an effort will be made to distil exactly what is intrinsically Latin American about these characteristics and what features are shared with other regions of the world; in other words, an attempt will be made to distinguish between traits of Latin America’s development that are universal or more widespread and those that are not. On the other, without setting aside the effort to arrive at generally valid observations, we will try to describe the nuances, typologies, and differences that shape the region’s diversity. However, limitations of space and the lack of certain types of data, as well as our own limitations, will make it impossible to provide equal coverage of all the different regions and countries.

Throughout this book we will also try to illustrate the ways in which different parts of Latin America compare with the rest of the world. The Latin American region is not part of what is regarded as being the “developed world.” None of the countries in the region has attained uniformly high enough living standards, educational levels, degrees of competitiveness, or levels of technological development to be ranked as a developed country. The failure to, as yet, take this giant step forward and the persistence of poverty and inequality have not, however, stopped the region from growing, from raising its population’s living standards, or (p.2) from improving its level of human development. Although some Latin American countries are still very poor and large segments of their populations are still on the sidelines of modern economic and social development processes, Latin America as a region has made large strides in bringing about very notable economic, social, and political changes that have placed it on a development path that has enabled it, as a whole and in some areas, to attain middle-income status on a global scale.

From the standpoint of the region’s production structure, and the above-mentioned progress notwithstanding, some traits have proved to be very long-lasting. Since the times of the Conquest, even while the region’s ties with the world economy underwent various changes and shifts, and even though some countries have managed to diversify their production structures and gain entry into international markets for manufactures and services, the bulk of the Latin American countries have not been able to leave their natural-resource-based production patterns completely behind them. This also implies that, above and beyond the fluctuations and differing circumstances seen in different product categories, Latin America’s pattern of trade specialization has held it back from gaining access to more technologically dynamic segments of the global market or segments in which the growth of demand is more robust. This pattern of specialization, along with the region’s markedly cyclical access to capital markets, also helps to account for the highly volatile nature of Latin America’s economic growth, which has undercut its development efforts.

This pattern of specialization, which clearly differs from the more capital-intensive and technology-intensive pattern of the developed countries and from the East Asian countries’ initially more labor-intensive pattern, but with increasing technological content, is not the actual reason why Latin America is not yet a developed region, but is instead simply a manifestation of that situation. Other countries and regions have been able to leverage their natural resource endowments in ways that have enabled them to bring about sweeping economic changes. With differing degrees of success at different stages in their development processes, the United States, Canada, Australia, New Zealand (a group of countries that we will call, using Maddison’s terminology, the “Western offshoots”) and the European Nordic countries provide examples of countries that have taken advantage of their natural resource endowments to place themselves upon development paths that have been more successful than those followed by Latin American countries. East Asian countries that have based their development strategies on their abundant labor supply, which shares some traits with some areas of Latin America, have been much more successful in achieving sustained economic growth in recent decades and in improving their population’s quality of life. The Latin American region’s limited success in terms of economic development and the insufficient diversification of production structures has made it especially difficult for it to sustain broad-coverage welfare policies.

This raises the question as to what factors have held Latin America back from making more radical changes in its economy and society and from doing more to improve its population’s quality of life. The answers to this question cannot be provided by economic analysis alone. Economic performance is the outcome of a complex constellation of social, cultural, and political relationships and of how those factors interact with the geographical setting. Within the realm of development theory, there is a long-standing debate about the role of institutions and (p.3) about the ultimate determinants of institutional development. The region’s social structures, the distribution of power and wealth, the role and strength of its elites, and the complex, often painful process of state-building (which in many cases has resulted in endemically weak nation states)—in combination with the legacy of colonial times and the economic and political difficulties that the newly independent states had in positioning themselves on the world stage—have all been decisive factors and all have something to do with the successes and failures of Latin America’s economies.

Latin America in the world economy: convergence and divergence in per capita GDP

Historical statistics on GDP trends in Latin America are quite limited and do not provide enough evidence to allow us to make categorical statements, especially in regard to the nineteenth and earlier centuries. Appendix Tables A.1 and A.2 offer a new historical data series for Latin America that has been constructed by drawing on our own estimates and on various sources, particularly ECLAC estimates for the period since 1950 and the statistics compiled by Angus Maddison for earlier years, which were in turn based on partial ECLAC estimates, those of other authors and estimates of his own. This new database, which uses the internationally comparable benchmarks developed by Maddison in 1990 as a point of reference, is expressed in international dollars at 1990 prices. The results differ—in some cases quite noticeably—from those provided in Maddison’s database. An overview is provided in Table 1.1.

While taking care in assessing these data, it can nonetheless be said that, over the past two centuries, per capita GDP in Latin America has fluctuated around the world average while going through three major phases: a decline between independence and about 1870 (although only relative to the world leaders of the industrialization process, which we will refer to here simply as “the West”), an upward trend in 1870–1980 (although with some ups and downs), and another decline since the 1980s. Thus, along with the European periphery, Latin America was able to join in the wave of modern economic development quite early on and to become part of what might be described as a “middle class” at the global level.

However, a comparison of Latin America with the West shows a long-run deterioration that was basically the result of two major downturns from which the region did not manage to bounce back. The gap between the Latin American region and the West first widened during the years before 1870 and then again during what has come to be known as the “lost decade” of the 1980s. On the other hand, Latin America has far outpaced Africa in terms of economic growth and continues to do so up to the present day. It also outdistanced Asia until the mid-twentieth century, but, since 1980, just the opposite has been true (and markedly so).

Recently, in an effort to understand the uneven economic growth rates of different nations, the concepts of a “little divergence” and a “great divergence” vis-à-vis the industrialized world have come into use. The Western economies underwent a major transformation as they transitioned from a pattern of slow

Table 1.6. Determinants of volatility, 1870–2008

Total volatility

Share of the first product

Average per capita GDP

Average growth rate

Terms of trade volatility

Argentina

6.9%

23%

5,129

1.6%

9.1%

Brazil

5.4%

54%

2,170

1.7%

14.1%

Chile

7.4%

40%

4,156

1.9%

12.7%

Colombia

2.9%

49%

2,320

1.7%

16.2%

Costa Rica

5.9%

53%

3,449

2.0%

12.9%

Cuba

11.6%

77%

1,866

1.9%

9.6%

El Salvador

6.7%

70%

1,994

1.3%

17.9%

Guatemala

7.9%

64%

2,613

1.7%

17.9%

Honduras

5.4%

43%

1,604

0.9%

14.3%

Mexico

4.8%

31%

3,500

1.8%

8.6%

Nicaragua

9.0%

40%

1,797

0.8%

19.1%

Peru

5.6%

29%

2,548

2.2%

9.3%

Uruguay

7.4%

38%

4,240

1.5%

14.1%

Venezuela

8.4%

63%

4,408

2.6%

16.9%

Correlation coefficient

0.441

0.013

0.045

0.062

Note: The correlation coefficient is calculated as the cross-correlation between the variables and total volatility.

Source: Authors’ elaboration based on Tables A.3, 1.7, and A.2

(p.4) (p.5) (p.6)
Latin America in the World Economy, 1810–2010

Figure 1.1. Population and per capita GDP growth rates, 1500–1820 and 1820–2008

Source: Table 1.1

economic growth between 1500 and 1820—during which these economies’ expansion was mainly accounted for by the growth of the population and, to a lesser extent, an increase in per capita GDP—to a different pattern, starting in about 1820, in which per capita GDP growth was a far more important factor than population growth (see Figure 1.1). During the first of these periods, the growth pattern of the “rest of the world”1 (which encompassed what is now regarded as the developed world, including Latin America) was entirely extensive and was slower than the pace achieved by the West, thereby giving rise to the “small divergence.” By the time that this process reached its end, the differences in per capita income levels were still small compared to the differentials that exist today. During the second period, although the growth rates of the “rest of the world” also rose, the increase in that group’s per capita GDP proved to be, in the long run, no more than a third as much as that achieved by the West. This was the origin of the “great divergence,” after which the gaps between these two groups’ levels of per capita income were very large indeed.

Ever since the time of independence, Latin America appears to have followed a growth pattern similar to that of the “rest of the world,” with the upswing in its growth rates being driven by the same factors as the rest: population growth accounts for some 60 percent of its total economic growth, whereas annual growth rates for per capita GDP in Latin America have been only three-fourths as high as the rates achieved by the West. Between 1820 and 2008, the gap between Latin America and the West widened from 0.8 to 2.7 times Latin America’s per capita GDP or, in other words, the region went from having a per capita GDP of slightly (p.7) more than one half of that of the West, to having one of slightly more than one fourth.

It is extremely difficult to talk with any certainty at all about the levels of per capita income during the colonial period, and the figures on population growth during that time are also highly debatable. If we go by Maddison’s highly speculative assumptions, it would seem that there was a substantial gap between Latin America and the West during the colonial period, although it did not increase to any significant degree during those years and, in Mexico, it may have been quite small at some point in time. During the early years of colonization, standards of living and, in particular, life expectancy at birth plunged. Then, however, the survivors of the Conquest and the colonizers saw somewhat of a recovery in their income levels, with the result that the gap at the end of the colonial period may not have been much greater than it was at its start.

In sum, while the West’s growth pattern was extensive and entailed fairly slow rates of increase, the gap between it and Latin America was sizeable but was not widening. When the West’s growth pattern changed to one marked by larger productivity gains, Latin America began to fall further behind and the gap widened to substantial proportions, even though Latin America’s pace of growth also picked up. Consequently, although the original gap and the legacy of the colonial period are subjects that attract a great deal of interest, the fact remains that new growth patterns emerged during the Industrial Revolution that radically changed the economic landscape and international relations. It would therefore be difficult to argue that the region’s more recent history is nothing more than a reflection of its colonial past.

The stages of development in Latin America

There is a major debate going on at the present time as to the accuracy of the statement that there was a long hiatus in Latin America’s development after the countries won independence as a result of the major conflicts and internal political instability that ensued, and that this situation did not change until the first wave of globalization2 finally pulled the region along in its wake. This debate will be examined in detail in the following chapter. As we will show, the general idea that the first decades following independence were not good ones for the region in economic terms relative to the countries that are today part of the developed world is still valid. It is also true, however, that, if we look a bit more closely, we see that there are significant differences, even across different regions within single countries. Be that as it may, between 1820 and 1870, the per capita (p.8) GDP gap between Latin America and the West swelled from 0.8 to 1.8 times the level of the former. While some countries and former colonies of other regions of the world (the “Western offshoots”) grew rapidly, and even though Latin America’s economies were not stagnant by any means, the fact remains that, in terms of the potential that seemed to be opened up when such natural-resource countries won their independence, the first decades of independence appear to have been a time when the countries missed an opportunity to match the pace of the more dynamic economies.

Between 1870 and 1980, in very different contexts and with some fluctuations, Latin America improved its global ranking, in contrast to the decline registered by the “rest of the world” up to the mid-twentieth century. What is more, Latin America’s share of world output climbed steadily, rising from 2.6 percent in 1870 to 5.2 percent in 1929 and to 9.5 percent in 1980 (see the last row in Table 1.1). Even so, the region was unable to narrow the gap with the West, which remained fairly stable during this period and actually widened somewhat during some years, especially between 1950 and 1973, when the Western economies marked up record-breaking growth rates during what was known as the “golden age” of capitalism. This was a time when the region was witnessing a population explosion and when the more developed economies in the region were having trouble finding a way to sustain their growth paths (see below). A more detailed examination of the situation shows that the gap with the “Western offshoots” widened but that it narrowed in relation to the industrialized countries of Europe up to 1929, while the opposite occurred after the Second World War.

This long period encompassed two completely different phases, however, not only in the history of Latin America but in that of the entire world. The first of these phases corresponds to the first wave of globalization and, in Latin America, a stage in which commodity exports were expanding. This first wave began to wane, in many ways, starting with the First World War and then collapsed altogether during the Great Depression of the 1930s. This also marked the beginning of a new phase of Latin American development which we will refer to as the era of “state-led industrialization.” At the international level, this phase began with a severe global economic crisis but ended with the emergence of a second wave of globalization starting in the 1960s.

Since 1980, Latin America has not only been lagging further behind the developed economies, but has also lost ground vis-à-vis the world average. While many nations, especially in Asia, have joined others on a rapid economic growth path, Latin America has grown at a substantially slower pace. As a result, the region’s share of world output slipped from 9.5 percent in 1980 to 7.8 percent in 2008.

Latin America’s economic performance has thus been marked by a certain duality, inasmuch as it has generally turned in an above-average performance (except during the first few decades after independence and the last few decades of the twentieth century and the first of the twenty-first), but, at the same time, has never closed the gap between it and the West. What is more, that gap widened further during these two phases of relative retrogression and, to a lesser extent, during the “golden age” that followed in the wake of the Second World War.

However, this somewhat disappointing pattern of lights and shadows is not a reflection of stagnation or of inertia. Since independence, the region’s per capita income has increased by a factor of 11, and, if we factor in its population growth, (p.9) then we can see that GDP has increased by a factor of 284, whereas the GDP of the West has expanded by a factor of 118.

This economic expansion has been coupled with thoroughgoing structural, institutional, and political changes, which have been brought about by both national and international factors that ultimately had a profound impact on the Latin American population’s lifestyles, culture, and quality of life.

All of this has occurred against a backdrop of deep inequalities within the region, which were already clearly apparent as early as 1820. Until about 1913, the degree of inequality among the Latin American countries increased as part of a tendency that, according to Gelman (2011), had begun at around the time that the countries were winning their independence. Argentina and Uruguay had high income levels from early on and, by 1870, Chile and Cuba had joined the ranks of the high-income countries.

Starting in the 1910s, however, this trend began to change as the high-income countries grew more slowly and a number of other countries, such as Brazil, Mexico, Colombia, and Venezuela, among the larger nations, and Costa Rica and Panama, among the smaller ones, began to grow more rapidly. As the growth rates of the countries that had been more successful in the nineteenth and early twentieth centuries declined and these new growth centers emerged, a process of income convergence among the Latin American countries took place. At the same time, however, there was a numerous group of low-income countries that continued to lag behind.

Events after 1980 also led the countries in different directions. While most economies tended to lag, some were quite strong: Chile, the Dominican Republic, Colombia during the “lost decade” of the 1980s, and Peru during the first decade of the twenty-first century. The result was that per capita GDP rates tended to converge up to 1990 and then began to diverge again.

A typology for an analysis of the Latin American countries

It is extremely difficult to find any typology for the Latin American countries that provides equally useful insights into their development process throughout the 200 years that have passed since independence. Certain characteristics were decisive during the colonial period but, as a consequence of economic growth, structural change, and social developments since that time, new traits or factors have taken on greater importance and capture the existing differences more clearly. Thus, one typology may be more informative in one period but may contribute less to the analysis in another. To complicate matters further, the information available for use in analyzing the different countries and their performance does not always provide full coverage of the different cases. The fact that some of the larger countries exhibit sharp differences from one subnational region to the next and the lack of disaggregated data constitute yet another difficulty.

(p.10) Nonetheless, some countries and regions have exhibited a number of specific features that have remained in place over time and that even today have retained a certain explanatory power.

As discussed by Cardoso and Pérez Brignoli (1979), the Latin American societies have been shaped by the interaction of three different societies that came together in the Americas: those of the pre-Columbian indigenous population, Europe, and Africa. The coalescence of these societies within different social and environmental contexts gave rise to three major types of colonial societies which may be understood as outgrowths of the dominant European society but which also developed specific traits of their own. Drawing upon the ideas of these authors, who in turn based their work on many other attempts to construct typologies in this area (Furtado 1976; Sunkel and Paz 1970; Cardoso and Faletto 1971),3 we will use the following criteria:

(a) The type of colonial power. The choice of this criterion has been a controversial one, especially in terms of its use to differentiate the colonies of highly mercantilist countries, such as Spain and Portugal, from the English colonies. While it is true that no former Portuguese or Spanish colony has yet become a developed country, there are many former English, Dutch, French, and Belgian colonies that have not reached that status either. While this factor is important, it is not as decisive as some authors, including Landes (1999), have described it as being. Others have rejected this characterization because they have seen it as attributing importance to the genetic make-up of certain population groups rather than to the specific qualities displayed by different societies over a given period of history. Clearly, however, a Hispanic heritage is not at odds with development. For the purposes of this analysis, the relevant difference is between the Spanish and Brazilian colonies.

(b) The type of market into which each society is most fully integrated. Here, a distinction can be drawn among export economies, economies that are subsidiary to export economies, national markets, and border or marginal zones. These various activities are not necessarily taking place in discrete sectors; they may overlap, although to different extents and in different ways, in each country or region.

(c) The main type of commodity, particularly in export activities: mining, agriculture, or forestry. In the case of agricultural products, there is an important difference between temperate and tropical climates, because of both the nature of the production processes in these two zones and the types of competition or complementarity that production in these climatic zones entails vis-à-vis buyer markets. Indeed, specific commodities give rise to different options for technological development and social organization, although this does not mean that there is a strict deterministic correlation between the types of resources that a country has and the path that its technological and institutional development will take, as argued by a number of highly influential authors in recent years (Engerman and Sokoloff 1997). It is, however, true that different products offer the possibility of different types of forward and backward production linkages. Export and competing markets also have different structures. This covers the entire (p.11) range, from products in which some of the countries of the region have monopolies or oligopolies (usually for a limited period of time), such as nitrates, coffee, or rubber, to those commodities (mainly tropical or subtropical agricultural goods) in which Latin American countries compete with regions with abundant, relatively inexpensive labor (Asia and Africa), to agricultural products in which the region competes with developed countries that are less rich in natural resources and have higher wage levels (wheat, maize, meat, wool) (see Lewis 1969 and 1978; Bértola and Williamson 2006). One factor that, in particular, is going to become increasingly important is the capacity of different economies to alter and diversify their export structure in ways that will increase value added. This, in turn, will determine what their trade balance will look like, based on the technological content of the goods and services that they produce and the factors determining the competitiveness of the products that they export and import. Nonetheless, the way in which the domestic market evolves (i.e., the way in which the structure of consumption and production changes) will also have a strong influence. In this respect, the progress made in terms of industrialization and the development of modern services will be the key indicators of production diversification.

(d) One pivotal aspect of this structure is the different ways in which the transition is made to the kind of wage-based labor market that is typical of modern capitalist economies and that is now prevalent throughout the region as well. Even though all the Latin American countries have converged toward these types of labor relations, the starting points for those transitions have been very different and have left a very strong mark on the various societies that is still glaringly evident in their labor relations and their positions in the international economy. Cardoso and Pérez Brignoli have identified three major types of transitions. The first type was made by the areas that we will call “Indo-European” regions, where indigenous and mestizo groups constitute a large portion of the population. These regions are, for the most part, located in what were the major centers of pre-Colombian civilization, and they became the pillars of the colonial structure, in which ranching and farming, indigenous campesino (peasant) communities and mining activities were all combined. Various forms of forced labor were still in use in these areas until well into the twentieth century. The second type of transition was seen in societies where persons of African descent have had a strong presence, which we will refer to as “Euro-African” regions. They are primarily located in tropical agricultural areas (although they are also found in some mining regions), where the importation of slaves, the development of a slave-labor-based economy, and the complex process involved in the abolition of slavery have been pivotal factors. Finally, there are the “Euro-American” societies. These societies are located in the temperate zones of the Southern Cone, where very few members of the original population survived and where European immigration has been the main factor in the growth of the population, or in enclaves within one of the other two types of societies.

(e) Finally, size is another important variable. Starting in the twentieth century, once the processes of social change, industrialization, and the diversification of production were in place and moving forward, the size of a country seems to have taken on particular importance because it influences the possible scales of production and the available opportunities for diversifying production. In the following (p.12) discussion, we will see how this factor can take on a significant explanatory value in relation to the development of different countries, especially during the period of state-led industrialization.

When all of these criteria are combined, they open up a wide range of possible sets of conditions and development paths. In fairly large countries, many of these differing traits exist alongside one another in varying combinations. This is especially the case in Brazil, Colombia, and Mexico. But even in smaller countries, such as Ecuador, a diverse mix of conditions can be seen, with tropical plantations prevailing along the coast and the structures typical of societies with a large contingent of indigenous peoples in the highlands. Consequently, any attempt to establish a one-to-one correspondence between these sets of characteristics and specific countries will inevitably be no more than approximate.

Nevertheless, when the criteria set out in (c) and (d) above are combined, a powerful typology can be constructed that captures a large part of the conditions existing in Latin America, especially up to the early decades of the twentieth century. Table 1.2 illustrates how the Latin American countries can be classified in terms of these criteria, with the help of some tailor-made definitions to help characterize a few particularly ambiguous cases.

From the standpoint of the societal traits and production structures of the different countries, they can be grouped, on a fairly reasonable basis, into three categories: (i) countries where the hacienda, the indigenous communities, and mining activities have predominated in primarily Indo-European societies; (ii) countries in which tropical plantations have been the predominant economic activity in what are for the most part Afro-American societies; and (iii) countries in which Euro-American societies based on temperate-zone agriculture or mining have predominated. The word “predominant” has been used repeatedly because, in each and every case, there is a mixture of these traits. Predominantly Euro-American societies can even be found in tropical regions (e.g., Costa Rica, the Antioquia and Santander regions in Colombia, the Venezuelan Andes and the tobacco-growing area of Cuba, whose production structures date back to colonial times).

Once into the twentieth century, however, as labor markets evolved, as income levels rose, and as domestic markets expanded, the differences between the first two groups seem to have become more closely related to the size of the economies involved rather than the differences in their initial conditions. In both cases, there continued to be a relatively abundant supply of low-cost labor. In analyzing the events of the twentieth century and the first decade of the twenty-first, we will therefore continue to use three categories: the category of the temperate-climate, newly settled Euro-American economies of the Southern Cone will remain the same, but the other two groups will be merged and then subdivided into the large and medium-sized countries (Brazil, Colombia, Mexico, Peru, and Venezuela) and the smaller economies.

Table 1.3 shows how, between the time that independence was won and 1913, but especially between 1870 and 1913, the disparities between different Latin American countries were on the rise, as illustrated by the coefficient of variation in per capita income levels. Between 1913 and 1990, the trend changed, and their income levels converged. However, since 1990 intraregional divergence has been on the rise once again. (p.13)

Table 1.7. Export concentration, 1870–1973

Share of total exports

Top export product

Top three export products

Country

1870–73

1910–13

1926–29

1949–52

1970–73

1870–73

1910–13

1926–29

1949–52

1970–73

Argentina

41%

21%

22%

7%

26%

74%

50%

56%

19%

46%

Brazil

53%

52%

71%

63%

29%

82%

77%

76%

78%

41%

Chile

52%

31%

46%

5%

64%

n.d.

34%

77%

7%

67%

Colombia

8%

45%

65%

74%

54%

14%

47%

82%

90%

69%

Costa Rica

86%

37%

61%

43%

37%

n.d.

69%

92%

74%

70%

Cuba

n.d.

71%

79%

81%

75%

n.d.

92%

92%

5%

90%

El Salvador

n.d.

76%

74%

83%

45%

n.d.

n.d.

n.d.

n.d.

62%

Guatemala

65%

69%

79%

77%

32%

n.d.

n.d.

n.d.

n.d.

51%

Honduras

n.d.

12%

44%

65%

50%

n.d.

14%

46%

73%

68%

Mexico

85%

22%

23%

19%

8%

91%

31%

49%

38%

18%

Nicaragua

n.d.

48%

54%

33%

24%

n.d.

56%

69%

1%

53%

Peru

33%

18%

34%

32%

18%

57%

36%

71%

56%

30%

Uruguay

35%

40%

33%

47%

36%

76%

69%

77%

78%

63%

Venezuela

42%

49%

69%

92%

n.d.

n.d.

n.d.

89%

94%

n.d.

Average

50%

42%

54%

52%

38%

66%

52%

73%

51%

56%

Source: Mitchell (2003)

(p.14)

Table 1.8. GDP growth explained by the evolution of the income elasticities of demand for exports and imports, 1870–2008

ε

π

ε/π

y

z

y/z

y *

y/y *

Argentina

1870–1913

4.7

2.0

2.3

2.8%

1.4%

2.0

3.1%

0.9

1914–1944

−2.0

−3.6

0.6

1.1%

0.6%

1.7

0.4%

3.1

1945–1980

1.0

3.4

0.3

1.8%

2.4%

0.7

0.7%

2.4

1980–2008

2.2

3.2

0.7

1.0%

2.0%

0.5

1.4%

0.7

1870–2008

1.9

1.9

1.0

1.5%

1.6%

0.9

1.7%

0.9

Brazil

1870–1913

3.1

23.3

0.1

0.2%

1.7%

0.1

0.2%

0.9

1914–1944

0.4

0.9

0.5

1.9%

3.0%

0.7

1.4%

1.4

1945–1980

16.4

1.8

9.0

4.1%

0.3%

11.9

3.1%

1.3

1980–2008

1.9

4.8

0.4

0.8%

2.5%

0.3

1.0%

0.8

1870–2008

2.4

2.5

0.9

1.6%

1.7%

0.9

1.6%

1.0

Chile

1870–1913

3.3

1.9

1.8

2.0%

1.3%

1.6

2.2%

0.9

1914–1944

−0.6

−1.0

0.6

0.9%

3.5%

0.3

2.1%

0.4

1945–1980

3.4

4.8

0.7

1.3%

1.4%

1.0

1.0%

1.4

1980–2008

3.7

1.6

2.3

3.0%

1.7%

1.8

3.9%

0.8

1870–2008

1.7

2.0

0.8

1.7%

1.8%

0.9

1.5%

1.1

Colombia

1870–1913

0.9

2.1

0.4

0.5%

1.6%

0.3

0.7%

0.8

1914–1944

0.8

1.3

0.7

2.6%

3.2%

0.8

2.1%

1.2

1945–1980

3.5

2.3

1.5

2.3%

1.5%

1.5

2.3%

1.0

1980–2008

2.6

2.5

1.0

1.7%

1.8%

0.9

1.8%

0.9

1870–2008

1.8

2.0

0.9

1.7%

1.9%

0.9

1.7%

1.0

Mexico

1870–1913

1.9

0.8

2.3

1.8%

1.9%

1.0

4.4%

0.4

1914–1944

−0.2

10.5

−0.0

0.7%

3.4%

0.2

−0.1%

−10.0

1945–1980

3.9

2.4

1.6

3.2%

2.1%

1.5

3.4%

0.9

1980–2008

2.5

6.5

0.4

1.0%

2.7%

0.3

1.0%

0.9

1870–2008

1.9

2.9

0.7

1.7%

2.4%

0.7

1.6%

1.0

Uruguay

1870–1913

3.7

3.2

1.2

1.0%

1.1%

0.9

1.3%

0.8

1914–1944

−0.1

0.1

−0.8

1.1%

2.7%

0.4

−2.1%

−0.5

1945–1980

0.7

2.3

0.3

1.7%

2.7%

0.6

0.8%

2.2

1980–2008

1.4

1.5

0.9

1.7%

1.9%

0.9

1.8%

0.9

1870–2008

1.2

2.1

0.5

1.2%

2.0%

0.6

1.1%

1.1

Venezuela

1870–1913

1.6

0.7

2.3

2.3%

2.1%

1.1

4.8%

0.5

1914–1944

2.0

1.3

1.5

5.1%

4.0%

1.3

5.9%

0.9

1945–1980

5.3

2.6

2.1

3.0%

1.4%

2.2

2.9%

1.0

1980–2008

1.1

68.8

0.0

0.0%

1.9%

0.0

0.0%

0.7

1870–2008

2.4

1.8

1.3

2.7%

2.2%

1.2

2.9%

0.9

Average (1870–2008)

1.89

2.18

0.89

1.7%

1.9%

0.89

1.7%

1.02

ε: income elasticity of demand for exports by foreign countries

π: income elasticity of import demand by Latin American countries

y: GDP growth rate

y*: GDP growth rate with balance of payments equilibrium, estimated as y=e/p*z

z: growth rate of each country’s “relevant world” (trading partners)

Source: Authors’ calculations

(p.15)

(p.16) The increasing divergence that arose between the time of independence and 1913 is largely a result of the strong growth experienced during that period in the relatively richer (Group 3) countries of the Southern Cone. Around 1913, this group of countries had achieved income levels that were quite close to the average for what we generally call the West, or what is today considered to be the developed world, while the Afro-American and Indo-American lower-income economies (Groups 1 and 2) were growing very slowly. By 1913, their income levels had fallen by approximately 30 percent relative to their richer Latin American neighbors.

The convergence of the Latin American countries that began around 1913 was a net effect of different factors. On the one hand, the growth of the economies of the Southern Cone began to flag, moving them away from the income levels existing in the West—at first at a moderate pace, but then, from the 1950s on (when the developed economies were experiencing a golden age), much more rapidly. Signs of any narrowing of this gap did not begin to reappear until the last decade of the twentieth century and the first decade of the twenty-first. The historical trends in Cuba have been even more adverse, since it has experienced an uninterrupted, long-term divergence with the industrialized world ever since the 1920s, moving from its ranking as the economy with the fourth-highest per capita income level in the region in 1913 to one of the lowest-ranking countries in this respect now. This trend has been in evidence both before and after the Cuban revolution.

Another important factor behind the convergence of the Latin American economies has been the strong performance of medium-sized and large countries other than those of the Southern Cone. This group of countries, after having lagged behind the industrialized world prior to 1913, began to reduce the income gap until the mid-twentieth century and then maintained relatively stable income levels vis-à-vis the developed countries during their “golden age.” They also substantially narrowed the gap between them and the richer Latin American countries from 1913 on and especially during the period of state-led industrialization. Fairly broad-coverage information for the smaller and poorer countries is available only from 1950 on. While these countries had average incomes levels that were far lower than those of the leading Latin American countries, they did manage to narrow the gap up until the 1970s, which also helped to reduce the coefficient of variation. The countries in that group that have been more successful over the long term have been Costa Rica and Panama.

After 1990, as noted earlier, the diverging trend seen in 1820–1913 made a reappearance, but it is still too early to tell whether this is the start of a lasting trend or whether it is simply the result of short-run circumstances.

Convergence, divergence, and integration into the world economy

Convergence, divergence, and volatility

There are two major interrelated factors that lie behind these trends of convergence and divergence. Some Latin American countries have experienced periods (p.17) of burgeoning growth (“economic miracles,” to use the term that gained popularity when Brazil embarked on its growth surge of the 1970s) and have reduced the income gap vis-à-vis developed countries, but have not been able to sustain this convergence. The other factor is the high levels of volatility displayed by all the countries of the region.

Some Latin American countries have grown very rapidly at some junctures in their history and have attained fairly high levels of per capita income, but, so far, as a rule, they have not been able to sustain those growth rates. Rather than closing the gap with developed countries, these convergence episodes have been cut short. This indicates that the Latin American countries have experienced “truncated convergence” (to use the term coined by Ocampo and Parra 2007), and have therefore alternated between convergence and divergence regimes relative to the leading countries (Bértola and Porcile 2006). Some of the “economic miracles” have lasted for quite a long time, as exemplified by Argentina in the three decades leading up to the First World War, Venezuela between 1920 and 1960, and Brazil and Mexico during the four decades that preceded the debt crisis of the 1980s. These growth phases have, however, usually been followed by severe crises which distance these same countries from the developed world.

The volatility of economic growth in the Latin American countries has been another typical feature. Experiences at the international level seem to indicate that when economic growth surges, its volatility increases as well. This may be due to the nature of international trade cycles, business cycles, population shifts and international migration, fluctuations in capital flows, and even the transition from one style and pattern of technological change to another. At the same time, when a higher level of economic development is reached, the degree of economic volatility tends to decline, but it does not disappear altogether, as the deep global crisis experienced since 2007–8 has clearly indicated.

Be this as it may, everything seems to indicate that the volatility of the Latin American economies exceeds the norm. Table 1.4 shows the level of volatility present in different groups of countries as measured by their income levels during the last half-century (since 1960). As we have seen, Latin America is a very heterogeneous region in this respect, as it includes low-, mid-, and high-income countries. Nevertheless, as a whole, the region still exhibits greater volatility than

Table 1.9. Historical indices of human development, 1870–2000

LA 7

LA 7

LA 12

LA 12

LA 16

LA 16

Central Countries

LA/Central Countries

GDPpc

logGDPpc

LE

LE (conv)

EDU

EDU (conv)

GDPpc

logGDPpc

LE

LE (conv)

EDU

EDU (conv)

GDPpc

logGDPpc

LE

LE (conv)

EDU

EDU (conv)

1870

0.02

0.33

0.07

0.03

0.06

0.52

0.29

0.12

28.7

63.5

24.2

21.3

1880

0.08

0.03

0.06

0.55

0.32

0.14

30.0

23.8

20.5

1890

0.02

0.38

0.09

0.03

0.07

0.57

0.36

0.16

30.3

67.2

23.5

19.8

1900

0.03

0.40

0.14

0.04

0.09

0.04

0.09

0.60

0.41

0.13

0.40

0.18

28.6

67.0

33.8

28.3

23.8

19.7

1910

0.03

0.45

0.18

0.05

0.11

0.04

0.10

0.62

0.48

0.16

0.43

0.20

34.5

72.6

37.8

30.5

24.6

19.9

1920

0.04

0.46

0.23

0.06

0.12

0.05

0.11

0.63

0.55

0.19

0.46

0.22

34.0

72.5

42.1

33.0

26.3

21.0

1930

0.04

0.48

0.26

0.07

0.14

0.05

0.13

0.66

0.61

0.22

0.49

0.24

32.1

72.4

43.4

32.8

28.8

22.7

1940

0.05

0.49

0.32

0.09

0.16

0.06

0.15

0.69

0.64

0.24

0.51

0.26

29.9

71.7

50.1

38.0

32.1

25.0

1950

0.06

0.53

0.43

0.14

0.19

0.08

0.18

0.72

0.74

0.32

0.54

0.28

32.5

74.5

59.0

42.7

35.9

27.7

1960

0.07

0.57

0.57

0.20

0.22

0.09

0.24

0.76

0.77

0.36

0.58

0.31

30.9

74.8

73.9

57.1

38.9

29.6

1970

0.10

0.61

0.63

0.24

0.28

0.12

0.33

0.81

0.79

0.37

0.65

0.38

29.8

75.5

79.9

64.0

43.2

31.4

1980

0.14

0.67

0.70

0.29

0.33

0.14

0.41

0.85

0.83

0.43

0.71

0.44

33.9

79.0

84.2

67.7

46.2

32.2

1990

0.13

0.66

0.75

0.33

0.41

0.19

0.51

0.89

0.86

0.47

0.75

0.51

25.4

74.5

87.4

71.0

54.3

37.6

2000

0.15

0.68

0.78

0.36

0.46

0.22

0.62

0.92

0.89

0.52

0.78

0.55

24.0

74.4

87.8

69.2

58.6

40.2

2010

0.18

0.72

0.83

0.43

0.50

0.25

0.67

0.93

0.93

0.63

0.84

0.66

27.3

77.0

89.8

68.4

59.4

37.8

Notes: LA7 includes Argentina, Brazil, Chile, Colombia, Mexico, Uruguay, and Venezuela.

LA12 includes LA7, Bolivia, Costa Rica, Cuba, Guatemala, and Paraguay.

LA16 includes Argentina, Brazil, Chile, Costa Rica, Cuba, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Dominican Republic, Uruguay, and Venezuela.

GDPpc: GDP per capita index calculated as GDPpc tk = (GDPpc tk- 100)/(40000–100), where GDPpc tk is the value of per capita GDP in year t for country k.

logGDPpc: Logarithmic transformation of the GDP per capita index calculated as logGDPpcI tk = (logGDPpc tk - log100)/(log40000-log100), where logGDPpc tk is the logarithm of per capita GDP in year t for country k.

EDU: Education index (average years of education) calculated as EI=Ed tk/16, where Ed tk is the average years of education in year y for country k.

EDU(conv): Education index (average years of education) using a convex function of achievement calculated as: EDUconv tk = (log16-log(16-Ed tk))/(log16), where Ed tk is the average years of education in year t for country k.

LE: Life Expectancy at Birth index calculated as: LEI tk = (LE tk -20)/(85–20), where LE tk is the life expectancy at birth in year t for country k.

LE(conv): Life Expectancy at Birth index using a convex function of achievement, calculated as: (log(85–20)-log(85-LE tk))/(log(85–20)), where LE is the life expectancy at birth in year t for country k.

Source: Bértola, Hernández, and Siniscalchi (2011); Table A.4

(p.18) any other group of countries with similar income levels. This higher level of volatility is not, however, correlated with higher GDP growth rates.

It is difficult to gauge how much of the sluggishness of the Latin American economies is associated with this factor. It is no easy task to factor in all of the problems that a high degree of volatility brings with it in terms of social, corporate, institutional, and political stability and the possibility of planning medium- and long-term investments. While innovation is closely related to iterative and cumulative synergies, the gradual accumulation of knowledge and emergence of innovation are seriously hindered by economic instability. This situation is compounded by what are, in general, procyclical fiscal and macroeconomic policies, which have tended to exacerbate, rather than dampen, other shocks that adversely affect production activity (Kaminsky, Reinhart, and Végh 2004; Ocampo and Vos 2008).

Some authors (Fanelli 2008) have described this feature of Latin American economic performance as “excess volatility” relative to the levels seen in other economies that are at a similar level of development.

One significant aspect of Latin America’s economic volatility has to do with its position in the international economy. While, since the Industrial Revolution, industrialized countries have specialized and positioned themselves in capital-intensive goods, and the Asian countries have relied on their abundant supply of labor, and while both of these groups have increasingly focused on the accumulation of technological knowledge, the Latin American countries have relied primarily on their natural resources as a means of positioning themselves in the global economy. But the supply and demand for these resources have been subject to ups and downs, and the prices that they bring have been extremely volatile. In addition, the fact that these countries’ production structures are concentrated in a relatively small number of commodities makes them more vulnerable to changes in demand and prices and has made it more difficult for these economies to adapt to changing circumstances.

The markedly procyclical nature of international capital flows to developing countries also contributes to this excessive volatility. This was pointed out by Triffin (1968) in respect of the first wave of globalization and was also identified during the second (Ocampo 2008b). As a result, when international trade has had positive growth effects, economic activity has been boosted further by capital inflows. On the other hand, when the international business cycle enters into a downturn, the effects of the slump in the demand for commodities and in commodity prices are amplified by the sudden stop and even the reversal of capital flows.

Table 1.5 shows the weighted average volatility for all of Latin America (the figures of individual countries are given in Appendix Table A.3). The data series are broken down by trend and cycle, and estimates are given for each of these components. It is interesting to note that volatility has not tended to decline in either of these respects. The level of volatility has fluctuated and was particularly high during the interwar period of the twentieth century, but there has been no downward trend since then. Table 1.6 shows that there is no clear-cut correlation between average income levels and the degree of volatility. Nor is there a very close correlation between growth rates and volatility; in other words, having (p.19)

Table 1.1. Per capita GDP, population, and GDP, 1500–2008: Regional averages and ratio to the world average

1500

1820

1870

1913

1929

1940

1950

1973

1980

1990

2008

Per capita GDP (dollars)

West

776

1,231

2,155

4,194

5,247

5,695

6,740

13,963

15,903

19,500

26,369

Expanded West

702

1,102

1,877

3,671

4,590

4,991

5,642

13,067

14,950

18,750

25,285

Rest of the world

538

578

602

859

924

1,073

1,092

2,064

2,371

2,711

4,900

Latin America

416

684

772

1,540

2,076

1,993

2,442

4,451

5,441

5,067

7,118

Rest (excluding Latin America)

544

575

599

820

865

1,003

962

1,804

2,038

2,453

4,670

World

566

672

880

1,538

1,789

1,958

2,108

4,083

4,512

5,150

7,614

Ratios

Latin America/West

0.54

0.56

0.36

0.37

0.40

0.35

0.36

0.32

0.34

0.26

0.27

West-Latin America gap

360

547

1,382

2,655

3,171

3,702

4,299

9,511

10,462

14,433

19,250

Gap/Latin America per capita GDP

0.86

0.80

1.79

1.72

1.53

1.86

1.76

2.14

1.92

2.85

2.70

Per capita GDP (world mean = 1)

West

1.37

1.83

2.45

2.73

2.93

2.91

3.20

3.42

3.52

3.79

3.46

Expanded West

1.24

1.64

2.13

2.39

2.57

2.55

2.68

3.20

3.31

3.64

3.32

Rest of the world

0.95

0.86

0.68

0.56

0.52

0.55

0.52

0.51

0.53

0.53

0.64

Latin America

0.73

1.02

0.88

1.00

1.16

1.02

1.16

1.09

1.21

0.98

0.93

Rest (excluding Latin America)

0.96

0.86

0.68

0.53

0.48

0.51

0.46

0.44

0.45

0.48

0.61

World

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

Population (millions)

West

51

126

208

339

375

401

434

553

577

612

695

Expanded West

75

175

268

424

479

520

566

720

756

801

891

Rest of the world

363

866

1,008

1,369

1,599

1,780

1,962

3,203

3,684

4,468

5,804

Latin America

18

22

40

81

107

130

165

308

360

442

580

Rest (excluding Latin America)

345

845

967

1,288

1,492

1,650

1,797

2,896

3,323

4,026

5,223

World

438

1,042

1,276

1,793

2,078

2,299

2,528

3,923

4,440

5,269

6,695

GDP (thousands of millions)

West

40

155

449

1,423

1,967

2,286

2,922

7,723

9,168

11,943

18,337

Expanded West

53

193

503

1,557

2,197

2,593

3,193

9,402

11,296

15,020

22,536

Rest of the world

195

507

619

1,201

1,519

1,910

2,137

6,613

8,734

12,114

28,438

Latin America

7

13

29

117

194

243

385

1,314

1,896

2,158

3,954

Rest excluding Latin America

188

492

587

1,076

1,313

1,655

1,721

5,224

6,774

9,874

24,392

World

248

700

1,122

2,758

3,716

4,503

5,329

16,015

20,030

27,134

50,974

Latin America GDP/World GDP

2.9%

1.9%

2.6%

4.2%

5.2%

5.4%

7.2%

8.2%

9.5%

8.0%

7.8%

Note: Values are presented in constant 1990 dollars.

“West” includes 12 Western European countries, Australia, Canada, United States, and New Zealand.

“Expanded West” includes 30 Western European countries, Australia, Canada, United States, New Zealand, and Japan.

Source: Own elaboration based on Maddison (2008) and Statistical Appendix, Tables A.1 and A.2

Table 1.2. A typology of Latin American economies

A

B

C

X

Y

Z

Indo-American

Afro-American

Euro-American

Large

Medium

Small

1. Subsistence agriculture and mining

1.1. With strong mining export sector

Chile

Y

Peru

Y

Mexico

X

Bolivia

Z

Colombia

Y

Venezuela

1.2. Without strong mining export sector

Ecuador

Z

Paraguay

Z

Guatemala

Z

El Salvador

Z

Honduras

Z

Nicaragua

Z

2. Tropical agriculture

Brazil

X

Colombia

Y

Cuba

Z

Dominican Republic

Z

Venezuela

Y

Panama

Z

Costa Rica

Z

3. Temperate-zone agriculture

Argentina

Y

Uruguay

Z

Chile

Y

Summary

Until 1930

1. A. (except Chile and Venezuela): Bolivia, Colombia, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Paraguay, and Peru

2. B and C: Brazil, Costa Rica, Cuba, Dominican Republic, Venezuela, and Panama

3. A and C: Argentina, Chile, and Uruguay

Since 1930

1. and 2. Z: Bolivia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, and Paraguay, Peru, Cuba, Dominican Republic, and Panama

1 and 2. X and Y: Brazil, Colombia, Mexico, Peru, and Venezuela

3. A and C and 2.C: Argentina, Chile, and Uruguay

Highlights: South of Brazil; North of Mexico; Peruvian and Ecuadorian coasts; Colombian Caribbean; Panama as a logistic settle; Central American countries have tropical agriculture; Costa Rica is Euro-American.

Source: Authors’ elaboration

a fast or slow growth rate will not make a difference in terms of the level of volatility.

Another aspect of volatility is the frequency and severity of financial crises, whether they are external debt crises, balance-of-payments crises (identified by sharp exchange-rate adjustments), or banking crises—or, commonly, a mix of them. The upper part of Figure 1.2 depicts the historical variations in the (p.20) frequency of financial crises. In every case, the peaks follow periods of hefty capital inflows that had their origin, as has been analyzed in a vast amount of the literature,4 in what are essentially international business cycles: the boom in external financing that followed upon independence, the boom that preceded the international crisis of 1873 (which marked the start of a long worldwide period of deflation), the Great Depression of the 1930s, the Latin American debt crisis of the 1980s, and the new series of crises in the developing world that began in East Asia in 1997, with the last two of these crises merging into what can be seen as a single long, drawn-out crisis. These crises enveloped almost all the Latin American countries (and, in some cases, all nineteen of them5) in one way or another. The boom of the 1880s also triggered an international financial crisis: the 1890 Baring crisis, whose international epicenter was Argentina, but whose impact in the region was more limited, with only Argentina and Uruguay being severely affected. Only two of the major international financial booms have not been followed by financial crises in the region: the boom that preceded the First World War and the one that gave way to the severe worldwide recession of 2008–9. In both cases, however, the booms were followed by regional recessions and in the first case by the abandonment of the gold standard, following in the footsteps of European nations. This was the most important harbinger of the collapse of that standard during the Great Depression of the 1930s.

The lower portion of Figure 1.2 shows how the composition of these crises changed over time, focusing on periods when this phenomenon was particularly intense. As the figure makes clear, debt crises have been the most common problem in Latin America since independence. Steep devaluations associated with balance-of-payments crises have been frequent since the First World War, and this situation was also the main factor behind the crises that occurred between the mid-1950s and mid-1960s, which were not preceded by a boom in external financing. Finally, the most recent type of crisis has been in the banking sector, and these crises have become increasingly frequent since the 1980s. As a result, since the 1930s, most crises have been “dual” crises (combined debt and balance-of-payments crises) and, since the 1980s, many of them have been triple crises (the above two plus banking crises). Actually, in recent decades, we should also add in a number of other dimensions, such as high inflation (which, in Latin America, has historically been closely correlated with balance-of-payments crises), balance-of-payments collapses, and, in fewer cases, domestic debt crises.6

There has also been a notable degree of convergence between external trade cycles and capital flows. Crises are usually triggered by sudden export collapses that occur in the midst of critical international conjunctures (1873, 1890, 1913, 1929, 1973, 1979, 1997, 2008) and that also lead to plummeting commodity prices, (p.21)

Latin America in the World Economy, 1810–2010

Figure 1.2. Economic crises of Latin America 1820–2008. A. Number of countries on a currency, external debt, or banking crisis. B. Number of country/years in crisis by period

Notes:

The definitions of crises according to Reinhart and Rogoff are the following:

Currency crisis: annual devaluation greater than (or equal to) 15 percent with respect to US dollar (or the relevant currency)

External debt crises: outright default on payment of debt obligations including principal or interest.

Banking crisis: bank run that leads to the closure, merging, or takeover by the public sector of one or more financial institutions. If there are no runs, the closure, merging, takeover, or large-scale government assistance of an important financial institution (or group of institutions) that marks the start of a string of similar outcomes for other financial institutions.

Source: Database for Reinhart and Rogoff (2009) kindly provided by the authors

(p.22)

Table 1.3. Latin American countries: Per capita GDP, 1820–2008 (in 1990 international Geary-Khamis dollars)

1820

1870

1913

1929

1940

1950

1973

1980

1990

2010

Argentina

998

1,468

3,962

4,557

4,342

5,204

7,966

8,367

6,433

11,820

Bolivia

2,045

2,604

2,695

2,197

2,987

Brazil

597

694

758

1,051

1,154

1,544

3,758

5,178

4,920

6,762

Chile

710

1,320

3,058

3,536

3,312

3,755

4,957

5,660

6,401

13,229

Colombia

607

676

845

1,589

1,868

2,161

3,546

4,244

4,826

6,982

Costa Rica

1,555

1,733

1,930

4,230

4,902

4,747

7,876

Cuba

695

1065

2,327

1,688

1,244

2,108

2,313

2,724

2,957

3,997

Ecuador

815

1,055

1,109

1,607

3,258

4,109

3,903

5,278

El Salvador

1,216

1,298

1,739

2,653

2,454

2,119

3,447

Guatemala

1,613

2,571

1,955

3,140

3,772

3,240

4,172

Honduras

1,544

1,195

1,353

1,715

1,971

1,857

2,464

Mexico

733

651

1,672

1,696

1,788

2,283

4,831

6,164

6,085

7,832

Nicaragua

1,694

1,328

1,564

2,813

2,095

1,437

1,889

Panama

1,854

4,068

4,824

4,466

9,198

Paraguay

1,569

1,419

2,015

3,218

3,281

3,819

Peru

840

1,024

1,892

1,895

2,289

4,001

4,248

3,008

5,844

Dominican Republic

1,071

1,982

2,403

2,471

5,361

Uruguay

2,106

3,197

3,716

3,536

4,501

5,034

6,630

6,465

11,706

Venezuela

460

570

1,010

2,813

2,879

5,310

9,788

10,213

8,313

9,434

Average

683

790

1,559

1,956

1,993

2,442

4,451

5,441

5,067

7,272

Average “West”

1,231

2,155

4,194

5,247

5,695

6,740

13,963

15,903

19,500

27,356

Weighted average by groups

Group 1

713

692

1,373

1,963

1,780

2,220

4,163

5,072

4,890

6,674

Group 2

588

727

906

1,270

1,351

1,855

4,134

5,392

5,054

6,935

Group 3

832

1,461

3,673

4,276

4,065

4,801

6,964

7,540

6,426

12,204

Medium-sized and large countries

1,071

1,426

1,551

2,035

4,379

5,585

5,307

7,193

Small countries

1,663

2,779

3,204

2,941

4,398

Ratios

LA average/“West”

0.55

0.37

0.37

0.37

0.35

0.36

0.32

0.34

0.26

0.27

Group 1/West

0.58

0.32

0.33

0.37

0.31

0.33

0.30

0.32

0.25

0.24

Group 2/West

0.48

0.34

0.22

0.24

0.24

0.28

0.30

0.34

0.26

0.25

Group 3/West

0.68

0.68

0.88

0.81

0.71

0.71

0.50

0.47

0.33

0.45

Medium-sized and large countries/West

0.26

0.27

0.27

0.30

0.31

0.35

0.27

0.26

Small countries/West

0.25

0.20

0.20

0.15

0.16

Standard deviation (LA7)

166

362

1,226

1,264

1,181

1,562

2,645

2,515

1,678

3,169

Coefficient of variation (LA7)

0.24

0.39

0.63

0.52

0.50

0.49

0.50

0.41

0.29

0.37

Standard deviation (LA19)

1,289

2,037

2,193

1,914

3,349

Coefficient of variation (LA19)

0.53

0.46

0.40

0.38

0.46

Notes

Group 1 includes Bolivia, Colombia, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Paraguay, and Peru.

Group 2 includes Brazil, Costa Rica, Cuba, Dominican Republic, Venezuela, and Panama.

Group 3 includes Argentina, Chile, and Uruguay.

The group of medium and large countries consists of Brazil, Colombia, Mexico, Peru, and Venezuela.

The group of small countries consists of Bolivia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panama, Paraguay, and Dominican Republic.

“West” includes 12 Western European countries, Australia, Canada, United States, and New Zealand.

LA7 refers to the seven largest Latin American economies.

LA19 refers to all Latin American countries in the sample.

Source: Table A.2 and 2.4

which in turn translate into trade deficits. More often than not, these crises coincide with a contraction in the supply of external financing, which is usually abundant when exports are on the rise.

Integration into the world economy

One possible hypothesis is that volatility is brought about by external factors: either fluctuations in external markets or fluctuations in the terms of trade of each country (and are then amplified by the procyclical behavior of capital flows, which unfortunately cannot be factored into the following exercise).

Table 1.5 traces the fluctuations occurring in those parts of the world that have an impact on Latin America (which we will refer to here as the “relevant world”) via its exports7 and its terms of trade, measured as an unweighted average for all the Latin American countries (these figures are also given in Appendix Table A.3). The first of these variables reflects the growth in demand, while the second reflects the impact of international price movements. It is important to note that Latin America experiences more volatility than its “relevant world” does (even though the Latin American countries themselves are part of that “relevant world”). However, the terms of trade appear to be the factor that transmits greater volatility to the region, although its impact varies from one country to another.

The figures do not provide any clear indication that fluctuations in the terms of trade have subsided as time has passed. What Table 1.6 does show, however, is that there is a fairly close correlation between volatility and export product concentration. The second column gives the percentage share of total exports represented by the country’s most important export commodity. Table 1.7 provides a broader range of information, including the shares of the three most important export commodities between 1870 and 1970–3. This provides irrefutable evidence of the extreme concentration in a very few product categories that has characterized the Latin American economies throughout their history. In the closing decades of the twentieth century, despite efforts to diversify exports, the vast majority of the countries of the region continued to rely on commodity exports and exports of natural-resource-based manufactures (see Chapter 5).

This pattern of specialization has been a subject of debate for many years. In the long tradition of the structuralist school of thought, the persistence of this pattern of production has been seen as the main reason for the region’s failure to grow more rapidly. While classical and neo-classical schools of thought have not regarded sectoral specialization as being a serious problem in terms of development, structuralist thinkers have emphasized that the growth of international demand and technological progress have strong sectoral biases and that, therefore, patterns of specialization have a decisive impact on a given production sector’s ability to boost productivity. If we look at growth trends over the last three decades, it becomes clear (p.23) (p.24)

Latin America in the World Economy, 1810–2010

Figure 1.3. Share of Latin America in world exports

Source:

Tena-Junguito and Federico (2011) from 1820 to 1940; OXLAD from 1941 to 1949; and ECLAC from 1950 onwards.

The data for Colombia 1865–1910 is extracted from Ocampo (1984). In the case of Cuba, the information comes from ECLAC until 1970 and IMF since 1990.

World total according to Maddison (1995) until 1992 and FMI for the rest of the period.

that the fastest-growing economies in the developing world have been those that have diversified their production the most and, in particular, that the larger the share of industrial production and of high-technology manufacturing exports, the higher a given economy’s economic growth rate has been (Hausmann, Hwang, and Rodrik 2007; Ocampo, Rada, and Taylor 2009).

Latin America’s share in world exports is illustrated in Figure 1.3. The first wave of globalization came at a time when world trade was based on an exchange of raw materials and food products for manufactures. At that point, Latin America was in a good position, given its pattern of specialization. Its share of world exports thus climbed from 6 percent at the start of the 1880s to somewhat over 8 percent in the years leading up to the Great Depression of the 1930s (or from 5 percent to 7 percent, if we exclude Cuba). The region topped this threshold for a few years following the Second World War, but this was more a result of the devastation that the Second World War left in its wake than a sign of any further progress by Latin America.

The collapse of the international division of labor that had characterized the first wave of globalization was followed, after the Second World War, by the predominance of intra-industry trade among developed countries and by growing protectionism against agricultural products and textiles coming from the developing world. In this setting, and given the biases generated by industrialization policies, Latin America’s share of world trade sank to somewhat less than 4 percent in the early 1970s (excluding Cuba),8 or, in other words, to a level three percentage (p.25) points lower than it had been during the boom of the 1920s. When the second wave of globalization began to open up more export opportunities for developing countries in the 1960s, and when Latin American countries began to shift their economic policies toward an emphasis on export growth in the closing decades of the twentieth century, the region was able to regain some of the ground that it had lost, but it was still far removed from the levels that it had reached during the first wave, with a share of about 5.5 percent in global trade in recent years, versus the 7 percent mark (excluding Cuba) that it had reached at the end of the first wave of globalization.

Trends in the terms of trade

One important factor has been the trend in the terms of trade for commodities relative to the trend in the terms of trade for manufactures (see Figure 1.4). When these trends are examined at the global level, it can be seen that they have varied a great deal in different periods (Ocampo and Parra 2010; Erten and Ocampo 2012). Starting in the late nineteenth century, but especially during the boom that preceded the First World War, the real prices of agricultural and mineral products trended upward. The disarray that prevailed in the aftermath of the First World War and, in particular, the worldwide deflation of 1920–1 and the 1929 depression turned this trend in commodity prices around, however. This change in the trend came about in stages (or as the downward phases in long-term cycles), rather than as a steady decline. The first strong reduction came in the 1920s and covered all commodities. The second took place in the 1980s and 1990s and was characterized by a sharp drop in the prices of agricultural goods and a more moderate one (which is not statistically significant) in mining products. As a result, between the decade leading up to the First World War and 1998–2003, the terms of trade for commodities (other than oil) fell by 60 percent, with tropical agricultural products being the hardest-hit and mineral products being affected the least. Real oil prices also fell, but that decrease came later than the drop in non-oil commodity prices (the 1930s and 1940s) and, although they also plunged in the 1980s, they maintained a substantial part of the ground gained during the two oil shocks of the 1970s. The commodity price boom that began in 2004, which was driven by demand from China and was stronger in mining and energy products than in agricultural goods, has prompted many people to think that the world may be returning to the patterns observed during the first wave of globalization, which worked to the benefit of commodity producers, but it is still too early to determine whether or not this is part of a long-term trend.

The deeper long-term downturn in tropical goods prices gives us reason to take a serious look at one of the versions of the Prebisch-Singer thesis about the terms of trade.9 This version refers to the existence of stark structural and institutional differences between the manufacturing sector in industrial countries and the production of tropical goods sectors in more backward regions. In the latter, (p.26)

Latin America in the World Economy, 1810–2010

Figure 1.4. Real commodity prices (1980 = 100). A. Agricultural goods. B. Total and metals

Source: Ocampo and Parra (2010) and Erten and Ocampo (2012)

labor tends to be in ample supply and the history of the labor-market institutions that exist in those areas has been one of forced labor and very weak unionization. Markedly oligopsonic markets are also a characteristic of international commodity markets. In contrast, manufacturing goods markets tended to be oligopolistic and, in industrialized countries, which have generally dominated those markets, workers have been better organized. This asymmetric market pattern has also been highlighted by Lewis (1969 and 1983), who has also underscored the fact that international migration has tended to be segmented into two different types of flows: European labor migrating to new settlement areas (including South (p.27) America), and Chinese and Indian workers who migrated to tropical zones and who were expressly barred from new settlement areas after a certain point by destination countries (as exemplified by, in particular, the “white Australia” policy). These asymmetries have also been incorporated into other typologies of Latin American development which have underlined the fact that, unlike the temperate-zone, new-settlement countries elsewhere in the region, tropical Latin American areas have had to compete in international markets that are dominated by countries with low per capita incomes and extremely low wage levels (Bértola and Williamson 2006).

The trends in the relative prices of non-tropical agricultural goods also fit in with another version of the Prebisch-Singer thesis, although only after a few corrections have been made. This version refers to the structural changes that occur in both production and demand as the development process moves forward. Demand shifts toward higher-quality products that can meet new needs and that entail an industrialization process that is followed by an expansion of the share of services sectors. In these situations, the long-term trend is toward a low income-elasticity of demand for commodities, which would make it seem difficult to account for the upswing in relative commodity prices experienced during the first wave of globalization. As explained by Rowthorn and Wells (1987), however, the trend in the income elasticity of demand for certain goods is not linear. As major changes occurred in income levels in Europe and the United States and as a meat- and wheat-based diet gained currency and was substituted for other, less nutritional products, the income elasticity of the demand for these goods came to be much higher than it was for others. The income elasticity for coffee underwent a similar, temporary upswing during the period when it was ceasing to be a luxury product and becoming a mass consumer good (Ocampo 1984: ch. VII). Once this “gastronomic transition” had been made,10 the demand curve began to display a low income-elasticity, as apparently occurred in the European countries after the 1920s. This trend was heightened by the agricultural protectionist policies introduced at that time, which are still in evidence today in the developed world. A similar situation to the one seen in the late nineteenth and early twentieth centuries is now being witnessed in the Asian economies.

The situation with respect to mineral products, including fossil fuels, is quite different. Many of these products are not renewable and are subject to strong supply constraints, although, of course, those restrictions are variable depending on the available technology and the cost associated with their extraction and transport from remote areas. These markets are also subject to strong rigidities, particularly after periods when investment has been low, which have an impact on prices for what are sometimes quite lengthy periods of time, since the lead time for investments can be quite long. This production sector is also more likely to be taken over by monopolies or oligopolies than (tropical or non-tropical) agriculture. In addition, the demand for these products is not inelastic. In fact, it can be (p.28) highly income-elastic during certain stages of development, as has occurred in China in recent decades.

In discussions concerning the impact of the terms of trade, a recurring issue is the influence of the double-factorial terms of trade, which take into account changes in relative levels of productivity. In any event, only a very small part of the decline in the agricultural sector’s terms of trade can be accounted for by the relative increase in agricultural productivity in the second half of the twentieth century.

The trade balance

One of the interpretations of the theory devised by ECLAC about the development of the Latin American countries, or of countries on the periphery of the world economy, places emphasis on the trend in these countries’ trade deficits, rather than focusing on the deterioration in their terms of trade (Rodríguez 2006: chs 3 and 5). The main idea is that the fundamental problem of commodity exporters is that, no matter how much effort they devote to export promotion, the income elasticity of the demand for imports will inevitably be higher than the income elasticity of the demand for their exports. This idea has been expressed by Prebisch, Singer, Seers, and others and has been taken up by Thirlwall, who sees it as simply being one of the determinants of convergence and divergence in income levels.11 Krugman has also drawn attention to this relationship, although he places more emphasis on the supply side. Nonetheless, whether the emphasis is placed on demand components (as in post-Keynesian theories) or on supply (as in neo-classical theories) or on the interaction between supply and demand (as in the evolutionary and neo-Schumpeterian models that focus on technological change), the basic idea is that economic development entails structural change and that, if that change is insufficient, there will be a long-standing tendency to run trade deficits and that, while subject to expansion and adjustment processes, the economy’s growth rate will be determined by its relative propensities to export and import.

Table 1.8 depicts these relations for seven Latin American countries in different stages and for the period 1870–2000 as a whole.12 The unweighted averages for this entire period, which are shown in the last row, indicate that estimates of relative long-term growth fit in very well with a simple model that explains growth based on the relationship between the income elasticity of demand for exports and imports. In almost all of the countries, this fit is very good over the long term, although it is less so for specific periods. This is attributable to the high volatility of the terms of trade and capital flows, which have a positive or negative impact on the relative growth rate during shorter periods of time. In the aggregate, Latin America’s per capita GDP relative to that of the West dropped from 36 percent in (p.29)

Table 1.4. Growth volatility, 1961–2008

Average growth rate

Standard deviation

Coefficient of variation

OECD

3.35

1.66

0.49

High income: non-OECD

5.86

3.21

0.55

Upper middle income

3.81

2.45

0.64

Middle income

4.69

1.83

0.39

Lower middle income

5.86

2.39

0.41

Low income

4.08

1.85

0.45

World income

3.64

1.51

0.42

Latin America and Caribbean

3.91

2.63

0.67

Source: Authors’ estimations using GDP growth rates (PPP) data from the World Development Indicators (World Bank)

(p.30) 1870 to 27 percent in 2008. This type of exercise may account for the overall drop in Latin America’s GDP from 36 percent to 31 percent. The remaining four percentage points of the decrease can be accounted for by other factors, such as a deterioration in the terms of trade, over-indebtedness, or even population growth.13

When looking at more specific time periods, it proves to be difficult to find time frames that fit all the countries. It is noteworthy, however, that in all cases the relationship between the countries’ growth and the portion of that growth that can be accounted for on the basis of trade elasticities (the last column of the table) is greater during the period of state-led industrialization (1945–80) than it is during either of the two phases of export-led development (1870–1914 and 1980–2008) and that this portion is less than 1 during the export development phases. This is in keeping with the interpretation of the period of state-led industrialization that will be presented in Chapter 4, which belies the “Black Legend” which the more orthodox literature has woven around it. The interwar period (1914–44) was much more varied, but exports did decline in four of the countries and showed less elasticity in all of them except Venezuela. Growth capacity, under this set of circumstances, depended on the countries’ ability to cut imports in order to keep their balance of payments in equilibrium.

Development and the environment

Human history can be seen as the story of how populations, economies, and the environment have interacted and of how knowledge, technology, and social organization have evolved as people strive to increase their capacity for change and for generating social well-being by drawing upon the endowments of natural resources that are available to them. Contemporary theorists have repeatedly disagreed as to whether the ecosystem places significant restrictions on economic development or whether humankind’s ingeniousness, its capacity for scientific and technological development, and its social inventiveness will, time and again, overcome what once seemed to be insurmountable barriers.

As we enter into the second decade of the twenty-first century, the prevailing view of the relationship between human society and the environment appears to be a highly critical one. Whereas, until a few decades ago, natural ecosystem services seemed to have been recycling the by-products of human activity and maintaining the water, nitrogen, carbon, phosphorus, and other cycles, there have been growing signs that the expansion of the human ecosystem relative to the natural one is placing constraints on the development pattern embraced by human society over the past few centuries. The accumulation of carbon dioxide and other greenhouse gases in the atmosphere and of nitrogen oxides in the atmosphere, water, and soil are signs that the ecosystem is becoming unable to recycle the vast amount of waste being generated by society (Solbrig 2006).

(p.31) Just as production capacity and the earnings derived from production are unequally distributed in today’s world, so are the different world regions’ contributions to this ecological imbalance. The United States, Europe, and Japan generate a demand for much larger amounts of natural resources (and, in the case of the last two, of food) than they can provide. Apart from the question of how much of an impact access to raw materials and foodstuffs from the New World had on today’s developed countries, Latin America has certainly geared its development pattern toward supplying these goods and has therefore given rise to quite the opposite pattern. The other side of the coin, however, is that, as a result, it has not attained the standards of living that it could have achieved given its natural resource endowments.

In the course of this protracted process of change in the world’s natural and social ecosystems, Latin America has been discovered and rediscovered time and again. When Europeans discovered America, this apparent natural paradise had long before been settled by age-old civilizations that had transformed it, wiped out species of animals, built up agrarian systems, and constructed cities. The changes wrought by the colonial powers transformed the region’s fauna and flora as they introduced new farming techniques, cleared land for crops, and started up mining operations. However, these changes were largely confined to coastal areas and, in the case of mining activities, a few other enclaves. The most influential changes were in people’s lives, with the decimation of the indigenous population, mainly as a result of the introduction of new diseases, being the most significant of all.

As new sets of circumstances arose in the aftermath of the Industrial Revolution and as the demand for raw materials grew, the fallacy that Latin America was, even during the nineteenth century, still a “virgin” continent with untapped resources persisted. As noted by Orihuela (2010), the Eurocentric historiography of the nineteenth century, even while it sometimes viewed the indigenous populations of the Americas sympathetically (although, more often than not, adopting a racist perspective), portrayed them as savages living in a natural state within a natural setting. They were thus also seen as a potential obstacle to the appropriation of natural resources and habitats by these civilizations.

During the stage when commodity exports were driving development, the environment was altered even more markedly by the expansion of export crops, widespread deforestation, the construction of entire cities, the development of the railroad and the building of roads, the expansion and diversification of mining operations and incursions into new and, until then, undeveloped areas that were less attractive areas for human settlement but were rich in natural resources such as nitrates, rubber, and guano.

During the period of state-led industrialization and the modernization efforts that went along with it, the main changes that were coming about were associated with urbanization, industrialization, the expansion of the electricity grid to power it, road construction, and the rapid expansion of the agricultural frontier. It was also during this period that the energy transition (i.e., a shift to an increased and more widespread use of fossil fuels) that the more developed countries had made during the preceding period spread to the majority of the countries of the region.

On the whole, the strong increase (although at unequal rates in the different countries) in the consumption of fossil fuels during the first half of the twentieth century picked up even more speed in the second half. This long-term upswing (p.32) did, however, undergo marked fluctuations as a result of the behavior of supply and demand. The impact that these factors had on the different countries was largely determined by international conditions but also depended on whether they were energy producers or importers. For energy-importing countries, the world wars cut deeply into their energy supplies. In both groups, the Great Depression of the 1930s drove down the levels of economic activity and household consumption, causing the demand for energy to subside as well.

One peculiar feature of the energy transition in Latin America was that the substitution of oil for coal happened very early on (both in terms of timing and in relation to the levels of GDP that had been reached). This process displayed major fluctuations, however, both in comparison to how it had occurred in industrialized countries and in terms of the speed with which it took place in the different countries of the region (Folchi and Rubio 2006).

Mining activity has climbed steeply in recent decades. The region’s continuous urbanization has begun to be reflected in ever-greater degrees of urban sprawl and environmental pollution. Meanwhile, the unrelenting expansion of commercial agriculture has led to an increasing use of fertilizers and a significant reduction in forested areas, owing in part to what is not always well-regulated logging. In fact, the region’s cities cover no more than 3 percent of its territory, and it is the changes that have taken place in agriculture, forestry, and the like that have had the biggest environmental impact. It is estimated that no more than somewhere between 55 percent and 70 percent of the region’s original forests are still here today. Deforestation has been rapid and, even though Brazil is the country in which the largest areas have been deforested, the actual rate of deforestation has been highest in Central America and the Caribbean. Nevertheless, and even from a strictly agrarian standpoint, the strength of environmental impacts in Latin America is below the world average and deforestation rates have been declining (Solbrig 2006).

Be this as it may, as long as existing patterns of development around the world and Latin America’s particular type of production-sector specialization remain as they are, the Latin American region’s natural resources will come under increasing pressure even as the region’s potential for extensive growth nears its limits. However, the entire ecosystem is showing signs of distress, and this will necessarily prompt determined scientific, economic, and social efforts to alter these development patterns. This, in turn, will lead to what will no doubt be largely unpredictable changes in the demand for natural resources and in the ways that those resources are tapped, used in production processes, and conserved. This will be a valuable opportunity for overhauling Latin America’s development patterns and production structures and for exploring new development paths that will enable the region to combine the use of its natural resource base with the incorporation of knowledge into all aspects of its economic activity.

From economic development to human development

Economists have been becoming increasingly dissatisfied with the use of per capita GDP as a measurement of development and living standards. Per capita (p.33) GDP may be an acceptable measurement of the resources that people can access in order to secure certain living conditions, but it is not a good yardstick for measuring peoples’ capabilities to achieve those living standards (Sen 1993).

This is why there has been increasing reliance on the Human Development Index in recent decades. The most widely publicized, and simplest, version of this index, regularly published over the past two decades by the United Nations Development Programme (UNDP), is composed of equally weighted measurements of per capita GDP, life expectancy at birth, and level of education (illiteracy and educational coverage).

In the following paragraphs, we will examine a few different options for constructing a historical human development index.14 The estimate that we will use here, which is based on the work of Bértola, Hernández, and Siniscalchi (2011), provides updates of earlier estimates of per capita GDP and life expectancy at birth, but the most important innovation is that we include a series of average years of schooling for the population in the different countries of Latin America (Morrisson and Murtin 2008). This allows us to overcome the difficulties involved in combining illiteracy rates with educational coverage15 (in fact, for this reason, in 2010 the UNDP started using this measure of educational achievement in its estimations of the Human Development Index). The country indices are presented in Table A.4.

Education and human capital

A distinction should be drawn between the concept of education as a dimension of human development, on the one hand, and the concept of human capital as a factor of production and a determinant of an economy’s level of competitiveness, on the other. In the first case, a more highly educated population is an end in and of itself, since education increases an individual’s capabilities, in the broad sense of the term, and enhances a person’s participation in social life. In the second case, however, we are talking about a means to an end, and a different one at that.

The situation in Latin America with respect to education is a contradictory one. On the one hand, determined efforts have been made to bring about substantive improvements in the population’s level of education. On the other, when viewed from a comparative vantage point, those efforts seem to have come too little and too late, and Latin America has been at a clear disadvantage relative to other regions.

Around 2000, Latin America’s population had completed an average of 7.1 years of schooling, whereas the populations of the four countries that have dominated the world scene over the last two centuries (France, Germany, the United Kingdom, and the United States) had completed an average of 12.5. This (p.34)

Table 1.5. Volatility of GDP and terms of trade: Latin America and its “relevant world”

1870–1913

1914–1944

1945–1980

1980–2008

TOTAL

Volatility of GDP—Latin America

Associated with the trend

2.4%

3.1%

1.8%

2.4%

2.9%

Associated with the cycle

3.8%

5.6%

2.7%

2.6%

3.9%

Total

6.2%

8.7%

4.6%

5.1%

6.8%

Volatility of GDP—relevant world

Associated with the trend

0.8%

4.0%

1.9%

0.6%

2.5%

Associated with the cycle

1.8%

5.0%

2.9%

0.9%

3.2%

Total

1.7%

9.0%

4.8%

1.5%

5.6%

Terms of trade volatility

Associated with the trend

3.3%

5.6%

4.6%

3.9%

4.8%

Associated with the cycle

6.2%

10.8%

8.1%

8.3%

8.9%

Total

6.1%

16.4%

12.6%

12.1%

13.8%

Note: The “relevant world” refers to GDP growth of trading partners (see main text).

The volatility of the trend and the cyclical component are calculated as the standard deviation of the growth rate as a percentage of the trend.

Source: Table A.3

(p.35) means that the region’s level of education as of the year 2000 was 59 percent of what those developed countries’ level was (see Table 1.9). This shows just how far behind Latin America is in absolute terms. If, however, we look at the region’s performance during the twentieth century, it becomes clear that it has made a great deal of progress. In fact, perhaps the most impressive figure is the one that measures the level of education at the start of the twentieth century, when the region’s population had completed, on average, just 1.5 years of schooling, which was only one fourth as much as the population of the developed countries.

The construction of indices to measure relative performance levels has been the subject of a great deal of debate. As will be discussed in greater detail later on, it has been argued that the function that best represents progress in terms of life expectancy is a convex one, since it becomes increasingly difficult to achieve one additional year in the population’s life expectancy as that variable increases and approaches physiological limits. A similar argument can be advanced in regard to education (i.e., that it should be a convex function, with each additional average year of education representing a greater achievement at the margin) especially since, among other factors, once they have reached adulthood, people need to work and to earn a living so that others can study rather than work. If this is the case, then, the relative performance of Latin America vis-à-vis developed countries is even worse. As shown in Table 1.9, the region’s level of educational achievement would then be equivalent to only 40 percent of that of developed countries.

Even the best-performing Latin American countries in terms of education had very low average levels of schooling at the start of the century: 1.8 years for Argentina and 2.4 for Uruguay in 1900, while the average for the developed countries was 6.4. Other newly settled regions, such as Australia and New Zealand, had levels similar to those of the other developed countries mentioned earlier. The available data indicate that this educational lag relative to developed countries arose during the closing decades of the nineteenth century, but it is likely that it was taking shape throughout that entire century.

Data for the broader sample of the sixteen Latin American countries for which information is available show that the point in time at which Latin America narrowed the gap with the developed countries the most was during the middle decades of the twentieth century (see also Astorga, Bergés, and FitzGerald 2005; Prados de la Escosura 2007). This effort continued, however, during the last two decades of the twentieth century and the first of the twenty-first.

These levels are obviously an average of differing levels from one country to another, and they tell us nothing about differences in quality or the distributional inequality of educational opportunities within individual countries. Inequalities across countries increased up to 1940 and then began to decline. Meanwhile, Frankema (2009) has shown how the universalization of education may not be coupled with any substantial improvement in its quality; in fact, there may even be a trade-off between coverage and quality. The data that he compiled do, in fact, indicate that Latin America’s performance would be seen to be less satisfactory if measurements of quality (such as the rate of school failure) were to be included in the analysis. The large lag that he identified in this respect in the early 1960s has, however, tended to narrow since then. Nevertheless, reading, mathematics, and science tests administered recently under the OECD’s Programme for International Student Assessment (PISA) point to similar trends, as the Latin (p.36) American countries’ scores are far below those of developed countries (Hanushek and Woessmann 2009).

Part of the explanation for this lies in the clear relationship between education and development. In fact, educational levels can be regarded as a dependent variable of the level of development, as educational levels change as an economy advances. The direction of causality would therefore be from economic growth to a rise in the level of education. Viewed from this perspective, the levels of education existing in Latin America would appear to be in line with its per capita income levels.

This is inarguably part of the explanation. It is very rare to find a country with a relatively low per capita GDP and a very high level of education. Cuba is a conspicuous exception to this rule which also clearly illustrates the difference between the role of education in developing human capabilities and in promoting an increase in human capital. The fact that the provision of services to meet educational and other social needs is dependent on the level of development attained by a given country is the backdrop for a well-known paradox: those countries in which the need for education, health care, and other social benefits is the greatest are those that have the least available resources to draw upon in order to meet those needs, while the richer countries can devote a larger share of their income to building welfare states. This also backs up the idea that, far from what many theorists contend, increased public social spending does not erode growth but rather boosts it (Lindert 2004).

Figure 1.5 shows that, while this is true, it is also true that Latin America stands out from the rest of the world in the sense that, for any given level of per capita income, its level of education is lower than that of other regions. In other words, Latin America has performed less well in terms of education but, at the same time,

Latin America in the World Economy, 1810–2010

Figure 1.5. Latin America and the West 1870–1930: Per capita GDP in 1990 PPP US$ (x-axis) and average years of education of population 15 and above (y-axis)

Based on Bértola, Hernández, and Siniscalchi (2011)

(p.37) appears to have achieved a better economic performance at a lower level of education.

The main explanations that have been advanced for this phenomenon have to do with social structures and power relations.

The education system that was set up during colonial times was designed to preserve and legitimize the existing social order. Education was a way of “civilizing” the indigenous peoples, eradicating their customs and beliefs, and indoctrinating them in the Catholic religion. Teachers were members of different religious orders (Franciscans and Dominicans, among others), and there were very few schools. Higher education was primarily for the European and Criollo (creole) elite and the children of the Caciques (Indian chiefs), and instruction was given almost entirely in Spanish and Portuguese.

A more modern system of education emerged in Latin America in association with other processes, the most important of which was the creation of independent states. This was accompanied by the appearance of new political parties, the emergence of entrepreneurs, and the arrival of internal and foreign immigrants. All of this fueled rapid progress in the creation of education systems, the gradual universalization of primary education, and the narrowing of the educational gap between men and women. And all of this, in turn, promoted social mobility.

As pointed out by Reimers (2006), however, these processes were all imprinted with a very basic division between those who advocated a democratic, inclusive form of education and those who crafted hierarchical, authoritarian social structures that went hand in hand with a high degree of social exclusion. The nineteenth century was a time of political infighting between different elite groups. Two factors or events signaled a change in the direction that the education system was taking. The first was the gradual shift away from Catholic schools and toward the secularization of the education system. This change was driven by groups of intellectuals associated with the independence movement who saw the Catholic-dominated system of instruction as being directly linked to the authoritarian social order and religious dogma that eclipsed the individual and the truth. The second important factor was this movement’s emphasis on primary education, whose provision it saw as a duty of the state. The second half of the nineteenth century saw the creation, especially in the countries that were among the first to achieve advances in development, of what would become national education systems that would lay the groundwork for the changes that were to occur in the twentieth century (Reimers 2006; Meyer et al. 1992).

In the course of the twentieth century, the countries gradually consolidated the national public education systems that had begun to emerge in the previous century, albeit at different paces in different countries. Public education gained ground, with a strong link being formed between schools and local governments. While the two opposing ideological camps referred to above held their ground, the conservatives gradually came to accept the state’s new role in providing education and in opening up access to it. Their ideological differences continued to exist, but now shifted into other spheres. Two factors should be underscored here. First, as noted earlier, the increase in the number of students in the primary education system was not necessarily accompanied by an improvement in its quality. Second, as the universalization of primary education ceased to be controversial, (p.38) the focus of this protracted ideological dispute turned to secondary and tertiary education (Reimers 2006).

As Reimers says, the period of rapid industrialization drove the expansion of the education system, not only because human capital came to be seen as a key factor in economic development after the Second World War, but also because of the emergence of the idea that development could be planned. Development agencies began to become important players in national financing and in education. The World Bank and later the Inter-American Development Bank (IDB) provided financing for infrastructure works, the construction of universities and of vocational and secondary schools, and the expansion of the primary education system. These efforts brought about not only an increase in enrolment in general, but also the spread of a common teaching methodology that was oriented more toward driving economic growth than toward promoting people’s integral development.

The uneven performance of education policies in the first half of the twentieth century had, by the 1950s, led to clear differences in enrolment rates. According to Frankema (2009), if we focus on the point in time at which a transition was being made toward education for all, we can, in an approximate way, identify three groups of Latin American countries. In the last three decades of the nineteenth century, primary-school enrolment rose most sharply in Argentina, Chile, Costa Rica, and Uruguay, although, somewhat surprisingly, less so than in the British colonies of Jamaica and Trinidad and Tobago. After winning its independence from Colombia in 1903, Panama joined the ranks of the countries that were first to expand primary education. In the 1920s and 1930s, gross enrolment rates started to climb more steeply in Bolivia, the Dominican Republic, Ecuador, El Salvador, Mexico, Brazil, Peru, and Venezuela. Some of the poorest countries in the region, such as Guatemala, Honduras, and Nicaragua, lagged behind, as was to be expected, and the upswing in enrolment rates did not pick up speed until the early decades of the post-Second World War period.

This classification appears to be a reflection of some important features of Latin America’s historical legacy. The countries which were the first to increase school enrolment rates are the countries that were on the periphery of the colonial system, where the Iberian colonizers had much less of an impact than they did in the central colonial territories. These countries appear to have been: (a) the most heavily urbanized ones, (b) the most ethnically homogeneous ones, with European immigrants being in the majority (Argentina, Uruguay), and (c) rural societies with comparatively lower levels of inequality (Argentina, Uruguay, Costa Rica). The countries with the most limited development were generally the most highly stratified rural societies that had the lowest levels of urbanization and a highly heterogeneous ethnic make-up that included a relatively small Criollo elite. Most of the Latin American countries were somewhere in between these two extremes and began to invest in mass public education in the early years of the twentieth century, particularly in the 1920s and 1930s.

Most of the expansion in primary and secondary education has been in the public sphere. While the enrolment rate in private schools is substantial, private education—especially at the primary level—has not become dominant. This is less true at the secondary than at the primary level, however, and still less true at the (p.39) university level, where, generally speaking, enrolment in private educational institutions is higher.

Secondary-school enrolment was much lower at that time, partly because the increase in primary-school enrolment was achieved by introducing a series of measures that worked to the detriment of educational quality, which in turn led to an increase in grade repetition rates that held students back from entering the next level of instruction. By the same token, for most of that century, the emphasis was on achieving literacy for all and on universalizing primary education, while scant attention was paid to the secondary and tertiary levels, with the idea of equitable access to education at all levels being reduced to its minimum expression.

Reimers’s views tie in with the neo-institutionalist approach, which, in its turn, has brought back the entrenched beliefs of the Latin American structuralist tradition. The underlying idea, then, is that, despite all of the Latin American republics’ efforts to universalize education, the region has not achieved the levels of education that it would be expected to have attained given its level of economic development, and this situation can hold it back from making more significant economic and social changes.

Proponents of the neo-institutional school of thought maintain that Latin America’s lagging progress has to do with the concentration of wealth and political power in elite groups. In support of their thesis as to the distribution of political power, they contend that, even though investment in education is highly and positively correlated with per capita income over time and across countries, there is a great deal of variation that cannot be accounted for by income differentials. Inequalities in political power, as expressed in the percentage of the population that has the right to vote, appear to be associated with lower literacy rates and less educational coverage (Engerman, Mariscal, and Sokoloff 2009; Lindert 2010).

Frankema (2009), along with other authors, has looked for a relationship between the development of the education system and the concentration of land ownership. The basic idea is that large landholders are not in favor of educating the workforce because they fear that education will build workers’ capacities and thus their political power and because they are more inclined to make use of unskilled labor than to use education as a means of boosting productivity.

Lindert (2010) has recently described how the concentration of political power, the concentration of wealth, and low levels of education are linked to one another. He also attempts to account for what he calls “education anomalies” (countries with higher incomes but worse education systems than others) by analyzing their tax systems and looking at how fervently opposed high-income sectors are to paying taxes to finance the cost of universal public education.

There is another line of thought that complements this interpretation. While acknowledging that the above-mentioned mechanisms are at work, the question can also be asked as to whether or not it is possible for a fairly uneducated population (i.e., one with a low level of human capital) to generate a high level of per capita GDP. This has to do with a production function in which the abundant factor is not labor, human capital, or physical capital but instead primarily natural resource endowments. Latin America’s traditional export pattern, in which the predominant component is made up of commodity exports that incorporate relatively low levels of value added but that nonetheless do generate economic rents, would be the main factor underlying its tendency to have a high level of per capita income relative to its endowment of human capital as measured by levels of education.

(p.40) This characteristic, which could serve as a good point of departure for a development process, could also inhibit such a process if those who have control over the countries’ natural resources also manage to gain political control and if the production pattern is oriented toward deriving profits from those natural resources rather than toward building the capacity for innovation based on the development of human capital and structural change in the production sector.

This line of reasoning does not point to a straightforward relationship between natural resource endowments and levels of education. On the contrary, this relationship will always be influenced by the patterns of economic and political power in a given society. For example, the countries in the extreme south of South America, which have a generous natural resource endowment relative to the size of their populations, succeeded in attracting immigrants and attaining higher wages and higher levels of social expenditure than other parts of the region in which the per capita natural resource endowment was not as great. Countries that are rich in natural resources but in which ownership is highly concentrated and large sectors of the population are uneducated and have little political power (owing, in part, to racial and cultural discrimination) may lag even further behind in terms of education.

Life expectancy at birth

The last ten generations of human beings have availed themselves of technical and physiological developments that were not within the reach of the many more generations that preceded them. The expression “technophysio evolution” is used to refer to the interaction between technological advances and improvements in human physiology, which is a synergy in which the whole is greater than the sum of its parts. The huge increase in the longevity of the world’s population has gone hand in hand with changes in height and weight. This technophysio evolutionary process may also involve genetic processes of natural selection, but during the past three centuries, the non-genetic factors have been particularly influential (Fogel 2009).

During the twentieth century, the mean life expectancy in twelve Latin American countries for which information was available jumped from 29 to 71 years (data for the early twentieth century for some of the countries with the lowest life expectancies are unavailable, so the actual mean was no doubt lower).

The nineteenth century does not appear to be one in which economic growth brought any appreciable improvement in this variable for most of the world’s population. Fogel contends that this type of indicator may provide a clearer picture of living standards than income-based measurements do because real wages may entail “bribery” in the sense that workers agree to work under conditions in which they are at greater risk of on-the-job (including fatal) accidents. Using these types of biomedical indicators, the industrialization process in England is a clear example of one in which the degree of inequality seems to have increased even though income distribution appears to have remained unchanged.

The twentieth century appears to be the point in time when most of the advances in this respect converged and when the majority of the population saw a decline in inequalities in life expectancy at birth.

(p.41) The reduction in mortality rates that raised life expectancy stems from four main processes: the improvement of public health systems, advances in medical theory and practice, improved personal hygiene, and higher income levels and living standards.

The way in which life expectancy at birth indices have been constructed has also been a subject of debate. While life expectancy at birth has risen enormously, some argue that there are certain biological limits which we are approaching and that each additional year of life expectancy for the general population is therefore becoming more and more difficult to achieve. This is why, rather than having a linear index, it makes sense to use a convex function that will reflect the increasing difficulty of attaining further increases in life expectancy. This would mean that advances at higher ages would be weighted more than advances in average life expectancy at lower average age levels (see Prados de la Escosura 2007).

Table 1.9 shows an index constructed on the basis of a maximum life expectancy at birth of 85 years (a value of 100 on the index). Developed countries would have a value of 89, while Latin America (12) would have one of 78. If we were to adopt the less optimistic view of advances in life expectancy, then both groups of countries would be much further from the maximum. What is most important, however, is that, if we use the second yardstick, then Latin America would not be at 88 percent of the level of developed countries, as shown in column 13, but at 69 percent, which would indicate the distance that it has to cover in terms of the effort needed to match the life expectancy achieved by developed countries.

Regardless of which index is used, it can be seen quite clearly from the lower portion of Figure 1.6 that Latin America’s main achievements relative to developed countries came between the 1930s and the 1960s, when the deployment of social policies was at its height. The closing decades of the twentieth century, on the other hand, appear to have been a time of stagnation or even of regression in relative terms, especially if the index built on the basis of a convex function is used.

Historical human development indices

As we have seen, indices of each of the components of human development can be built in different ways. By combining two alternative versions of each indicator, we can construct eight alternative aggregated indices. To simplify, in Figure 1.7 we show three indices that reflect the two extremes and one intermediate one.

IR3 is the index that best reflects the values and criteria traditionally used by UNDP, although the reader should bear in mind that the information that we have on education (and that used most recently by UNDP) is completely different from traditional estimates (based on school attendance and illiteracy). This index uses the logarithm of per capita GDP and does not alter either the level of education or life expectancy at birth. The result indicates that Latin America has progressively narrowed the gap between it and developed countries, although, by around the year 2010, it was still at just 75 percent of the developed-country level. It also shows that more ground was gained during the middle decades of the twentieth century, particularly between 1930 and 1960.

If we use the IR2 index instead, which does not alter per capita GDP but does recalculate the indices for years of education and life expectancy at birth to reflect the fact that it becomes more difficult to make further advances as one moves up (p.42)

Latin America in the World Economy, 1810–2010

Figure 1.6. Latin America: Historical HDI relative to developed countries. A. Average years of education. B. Life expectancy at birth. C. Per capita GDP

Note: Developed countries include Germany, France, United Kingdom, and United States.

Source: Table 1.9

(p.43)
Latin America in the World Economy, 1810–2010

Figure 1.7. Historical human development indices 1900–2000: Latin America (LA7) and developed countries

Note: See Table 1.9 for details. Latin America (LA7) includes Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela.

Developed countries include Germany, France, United Kingdom, and United States.

Relative Index 1 (RI1): Geometric mean of per capita GDP, and life expectancy at birth and education indices (GDPPCI, LEI, EI).

Relative Index 2 (RI2): Geometric mean of GDP per capita index, and the indices of convex function for life expectancy at birth and education (GDPPCI, LE-con, EI-con).

Relative Index 3 (RI3): Geometric mean of log GDP per capita, and life expectancy at birth and education indices (logGDPPCI, LEI, EI).

Source: Authors’ calculations based on Bértola, Hernández, and Siniscalchi (2011)

the scale, we see, here again, that the greatest progress in relative terms was made between 1930 and 1960. After that, the rate of advance was slower and then came to a halt in the 1980s, though resumed the upward trend in the first decade of the twenty-first century. In this case, Latin America’s ranking on the Human Development Index has been in recent decades slightly over 40 percent of that of developed countries. The IR1 index, which does not alter the original values at all, reflects a trend similar to that seen with the IR2 index, with the difference that the relative value amounts to slightly over 50 percent of that of developed countries.

Clearly, none of these indices provides an entirely accurate picture of the situation. Each one reflects different aspects of human development trends, and each is influenced by what we think may happen in the future, as in the case of life expectancy at birth. Taken together, they provide us with a more nuanced view of the complex social and economic development process in Latin America, which, starting from an initial position in which it lagged extremely far behind the developed world, began to close that gap in the middle decades of the twentieth century (with varying lags in different countries) and then came to more or less of a standstill in the final decades of that century.

(p.44) Inequality

It is well known that Latin America is the region with the highest levels of income inequality in the world, which is why there has been so much interest in inequality studies in recent years. One of the main questions to be asked is: To what extent does the existence of inequality help to account for the region’s relative backwardness? The other is: To what extent is this characteristic a result of the region’s particular style of development?

After the Second World War, the tradition of research led by Simon Kuznets focused on the impact of economic growth and structural change on levels of inequality. Although the relative levels of productivity in urban and rural sectors played a predominant role in his analyses, he also devoted a great deal of attention to non-economic (at least in a strict sense) factors such as demographic, social and political changes linked to industrialization and to the development of welfare states.

In conventional economics, inequality has primarily been discussed in terms of income distribution. Viewing growth as a process of resource allocation, the main idea is to look at what kinds of price movements are experienced during processes of market integration and de-integration. The link between globalization and inequality has sparked a great deal of interest, with the focus being on the impact that initial endowments of factors of production (i.e., the relative abundance of land, capital or labor) have in terms of inequality (O’Rourke and Williamson 2006).

While this tradition continues to exist, the question of the kind of impact that inequality has on long-term growth has become one of the most dynamic areas of theoretical research. According to conventional economic thought, human capital formation is the main channel through which inequality affects the growth rate: the more equal a society is, the greater its rates of human capital formation and growth will be. Other channels for inequality’s negative impact on growth would be unequal access to financial resources and social and political instability.

The theoretical discussion has increasingly focused on determining which institutions promote inequality, both from the standpoint of income and wealth distribution and in terms of political power (North, Wallis, and Weingast 2009; Acemoglu, Johnson, and Robinson, 2005). From this vantage point, market forces and resource allocation play a secondary role.

The concept of human development has also been a focus of attention in this debate. In order to understand the dynamic of economic and social development, we have to look not only at the distribution of income and wealth, but also at the distribution of human capabilities.

Most of the research done on and in Latin America in the 1950s, 1960s, and 1970s (with, in the second case, the structuralist school of thought leading the way) underscored the importance of the oligarchic aspects of Latin American development, which were manifested in the concentration of political power, wealth, and income in elite groups of landowners and financiers who controlled labor relations and trade. While the winning of independence disrupted the political order, opening the way for chaos and making it difficult for the state to protect property rights and to maintain security, what we now call the first wave of globalization was associated with the consolidation of the state’s political power (p.45) and a heavy concentration of wealth, along with a stauncher defense of the elites’ property rights. At the same time, labor relations were changing, but they remained clearly subordinated to the interests of the elites and, in fact, became subject to old and new forms of extra-economic coercion typical of pre-capitalist forms of organization. Thus, the problem was not only the heavy concentration of land ownership, which was in itself excessive by international standards (see Frankema 2009: ch. 3), but also the continuation of various forms of labor subordination that curbed labor mobility. The countries that were relying on immigration from Europe were the conspicuous exception to the latter rule.

These new elements had their roots in the region’s colonial legacy. These new types of relations played out in different spheres, however, owing to the interaction of powerful international forces (the Industrial Revolution, the independence of the United States, the outcome of the Napoleonic Wars) that influenced the liberal reform movement in Latin America. These liberal reform processes took place in varying settings, however, due, as we have seen, to the combination of growth in the areas that played a central role during the colonial era, of the expansion of tropical crops that were cultivated with the use of large contingents of slave labor, and of advances in pushing back the frontier with the help of immigrants from Europe (Cardoso and Pérez Brignoli 1979).

Neo-institutional theorists have revived this older tradition of research (although they generally ignore the older Latin American structuralist literature), arguing that the institutions set up by the colonial powers immediately following colonization created a long-term equilibrium situation marked by sharp political and economic inequalities, sluggish human capital formation, and slow economic growth. Different authors put emphasis on different factors—from the role of the region’s political and cultural legacies (North, Summerhill, and Weingast 2000), to its natural resource endowment and population density (Engerman and Sokolof 1997 and 2001) or the balance of social and political forces (Acemoglu, Johnson, and Robinson 2003; Robinson 2006)—but they all agree as to the decisive role played by early colonial structures. This school of thought has had a strong influence, not only in the academic world, but also in major international organizations (see, for example, World Bank 2004).

These views have recently come under fire. Coatsworth (2008) contends that the root causes of Latin America’s backwardness should be sought in the period between 1770 and 1870, when the Latin American economies missed the opportunity to engage with the Industrial Revolution and bring about one of their own. In his view, the local elites did not become stronger in the colonial era but only later, toward the end of the nineteenth century, and—contrary to what the neo-institutionalists contend—that reinforcement would not have been possible had it not been for economic growth. Viewed from this standpoint, the concentration of economic power was not an adverse factor but instead one that fostered development.

Unlike the advocates of earlier traditions, the neo-institutionalists looked to local conditions as the sole explanation for the region’s development path and inequalities, while turning away almost entirely from an analysis of how those inequalities were reproduced at the international level and how they influenced the existence of inequalities at the national level. Similarly, the emphasis on colonial institutions has diverted attention from research into how institutions have been altered by their interaction with national and international processes of (p.46) change, with the most significant of those processes undoubtedly being the Industrial Revolution and the ensuing succession of economic growth impulses, technological change, and structural and social transformations (Bértola 2011).

There appears to be broad agreement that the degree of inequality in Latin America (both across and within countries) increased during the first wave of globalization. Proponents of this view do not deny that sharp inequalities were inherited from the colonial era but regard those conditions as feeding into the inequalities that arose during the first period of globalization. With 20 percent of the population being Caucasian (many of whom were very privileged), 25 percent slaves, a large mestizo group subject to multiple forms of discrimination, and an indigenous population subject to differing types of servitude, and given the strength of the colonial powers, sharp inequalities in terms of both economic opportunities and civil rights took shape and reinforced one another.

As we will see in Chapter 3, the increasing inequality experienced during the first wave of globalization was not confined to the impact of market forces. For the most part, it involved a redistribution of wealth and institutional changes that consolidated a particular configuration of ownership, wealth, and political power. Thus, far from being a purely economic phenomenon associated with an adjustment to a new type of equilibrium situation in the wake of the revolution in transportation, the increase in inequality during this period was closely linked both to the patterns inherited from the colonial past and to their interactions with new economic, social, and political forces.

The period of state-led industrialization had differing results in terms of equity in different countries. In those that developed some form of welfare state, this was a time of declining inequality. This is the case of countries such as Argentina, Chile, and Uruguay, which combined a relatively poor economic performance during this stage of development with advances in terms of improved equity, until this progress was cut short by the advent of bloody military dictatorships. In other countries with very large domestic markets and highly segmented labor markets in which a large percentage of the population was made up of the descendants of slaves or of mestizo campesinos (mixed-blood peasants) or indigenous groups, the industrialization process heightened the concentration of wealth and, even among wage-earners, led to a growing polarization of income levels. Brazil may be regarded as the epitome of this type of experience. In still others, these kinds of phenomena were in evidence until the industrialization process was quite far along, but at some point distribution began to improve, particularly as the surplus of rural labor began to shrink and the effects of the development of the education system began to be felt. Mexico, Venezuela, and Colombia are some of the main examples of this type of pattern. In the long run, a significant effect of industrialization and of the urbanization that went along with it, in conjunction with widely varying types of agrarian reforms, was the erosion and ultimate elimination of long-standing forms of servitude that had existed in rural areas, in particular.

There is a broad consensus, as well as detailed information, about how the market reforms of the late twentieth century led to a significant increase in inequality and about the association between this increase and deregulation, the destruction of state capacity, and de-industrialization. In the Southern Cone, the military dictatorships that seized power played a major role in the early phases of the reform process, which they accompanied with a systematic repression of the (p.47) attempts made by the people to organize, with one of the results being substantial reductions in real wages. More generally, however, the economic crises that took place had a powerful negative distributional impact, while the restructuring of production generated a demand bias for skilled labor that also had an adverse effect on income distribution. In the first decade of the twenty-first century, however, the trend in distribution improved and inequality declined. The reasons for this are still open to question, but it is clear that at least two types of policies have had a broad impact: the improvement of the distribution of educational opportunities had a cumulative effect, and the newly designed system of social assistance succeeded in reaching the poorest sectors of the population.

A comparison of today’s Latin American societies with those that existed when the countries of the region were just winning their independence clearly shows that the distribution of wealth (which is now based more on capital than on land) continues to be highly unequal and that income distribution is possibly even worse than it was then. On the bright side, however, after an initial period during which the situation deteriorated, educational opportunities and access to health care have since improved. Even more importantly, slavery has been abolished, along with almost all of the various forms of servitude that were typically found in rural areas. These latter developments represent significant advances in terms of improved equity, although their implications for providing more equal access to the full exercise of citizens’ rights are still not entirely evident.

Notes:

(1) We exclude “the larger West,” which aside from what we call “the West” (the leaders of the industrialization process, made up of the major European powers and the “Western offshoots”), also includes other Western European economies and Japan, which joined the ranks of the leaders later on.

(2) The terminology used here is in line with the recent tendency to refer to the worldwide economic expansion of the late nineteenth century and early twentieth century as the first wave of globalization. Some authors prefer to use this term to refer to the stage marked by the conquest and colonization of the Americas during the sixteenth century. But if “globalization” is understood as the process of market integration, then it would not apply to the period prior to the late nineteenth century, which involved a process in which political and military, rather than economic, forces prevailed.

(3) For a discussion of these typologies, see Bértola and Williamson (2006).

(4) See, particularly for Latin America, Bacha and Díaz-Alejandro (1982), Marichal (1989), Stallings (1987) and, for the more specific case of the debt crisis of the 1980s and the years leading up to it, Devlin (1989). For the global situation, see also the now classic work of Charles Kindleberger (a recent edition can be found in Kindleberger and Aliber 2005) and the more recent analysis of Reinhart and Rogoff (2009), which is the source of the data used to construct Figure 1.2.

(5) From the 1960s on, the figure does not include Cuba and thus covers eighteen, rather than nineteen, countries.

(6) These are the different dimensions covered in the analysis of financial crises presented by Reinhart and Rogoff (2009).

(7) The “relevant world” for each Latin American country has been identified on the basis of annual variations in the GDP of each of the destination countries for its exports, weighted according to the share of total exports that it represents from year to year.

(8) The data for Cuba for the 1970s and 1980s have not been included in the figure because they were inflated by that country’s trade agreements with the Soviet Union and the Council for Mutual Economic Assistance.

(9) For a discussion of the various versions of the Prebisch-Singer thesis, see Ocampo and Parra (2003).

(10) The “technophysiological evolution” (Fogel 2009) that has resulted in significant increases in longevity and changes in body characteristics during the last three centuries, which are unquestionably related to changes in diet, will be discussed below.

(11) See also Bértola and Porcile (2006) and Cimoli and Porcile (2011).

(12) Strictly speaking, this exercise should include relative export and import price indicators or the real exchange rate, but information on these variables is not available for this length of period; in addition, these variables may not necessarily reflect changes that have a significant impact over the very long term.

(13) It is important to remember that the “relevant world” for Latin America in this exercise is not the same as it is for the West’s per capita GDP as shown in Table 1.1.

(14) The first attempt to build a historical human development index for Latin America was made by Astorga and FitzGerald. This work was published as an appendix in Thorp (1998a) and later was revised by Astorga, Bergès, and FitzGerald (2005). Later, Prados de la Escosura (2007) presented a different version of this index, and Bértola, Camou, Maubrigades, and Melgar (2010) then introduced further changes for the Southern Cone countries and estimated distributional inequality in two cases.

(15) See Bértola, Hernández, and Siniscalchi (2011).