State-led Industrialization 189
question to answer, since most of the international comparative studies that have
been done cover more recent periods. Be that as it may, the evidence that will be
presented in the next chapter indicates that the depth of the financial market in
most of the Latin American countries was comparable in 1980 to that of other
countries at a similar level of development. However, Brazil's and the Southern
Cone countries' greater propensity to high inflation had an adverse impact on
financial development. This ended up eroding Argentina's leadership position in
this area and, to a lesser extent, that of Brazil, while the countries that were less
prone to inflation continued to deepen their financial sectors in the aftermath of
the Second World War (Goldsmith 1973). Brazil's invention of financial index-
ation in the mid-1960s and this system's adoption by other countries was a way of
counteracting the effect of rampant inflation.
The main macroeconomic problems that countries ran up against during the
period of state-led industrialization were thus the waves of balance-of-payments
crises that erupted during the years of external strangulation (from the mid-1950s
to the mid-1960s) and those that arose following the first oil shock in 1973. Fiscal
deficits and inflation were recurring problems in Brazil and the Southern Cone
countries, but they did not spread to the rest of the region until the end of the
period of state-led industrialization. In the case of fiscal deficits, this was the result
of the first boom in private external financing that the region was to witness in half
a century, while the inflation that then swept the region was part of a worldwide
phenomenon.
An overall economic balance
How should these ups and downs in Latin America's economic performance
during the state-led industrialization process be interpreted? Were the failures
of the Southern Cone countries, which fell further behind developed countries
during this period, a harbinger of the crises that the rest of Latin America was to
face in the 1980s? Or, on the other hand, would it be valid to say that, if it had not
been for the debt crisis, the state-led industrialization model could have been
redirected in a way that would have supported a continuation of rapid economic
growth? How should we approach, more generally, the idea that Latin America
has, throughout its history, as noted in Chapter 1, been plagued by episodes of
"truncated convergence," or, in other words, episodes of rapid growth that are
then cut short by others in which the region loses ground once again?
The above analysis indicates that the particular ways in which structural change
and industrialization took place in the region permitted a significant degree of
technological learning to take place and, most importantly, enabled the region to
achieve the highest rates of GDP and productivity growth of its history. But, as we
have noted, this process was also marked by setbacks in traditional export sectors
and by the limited development of national innovation systems, which held the
region back from engaging in a deeper, more sweeping process of structural
change. The set of social, political, and distributional arrangements involved in
the expansion of the domestic market faced increasing tensions as the region
proved unable to take the industrialization process further. One of the reasons for
this was the increasingly staunch opposition to industrialization put up by
190 Economic Development of Latin America
traditional export sectors; another was the confrontations with the working class
that were triggered by this process, especially in what were initially the most
advanced economies: those of the Southern Cone. Resistance to a more thorough-
going form of regional integration that would have covered competitive imports
was another obstacle. It can even be said that, with the conspicuous exception of
Brazil, the coalition of industrialists never grew strong enough to take the indus-
trialization process to the next level, in contrast to the dynamic industrialization
push that the more successful East Asian economies were making. This situation
was compounded by the structural volatility of an export sector that still relied
heavily on commodities, which exposed the system to frequent external crises, and
by the financial volatility associated with the region's return to international
financial markets in the 1970s. The region had, of course, a long history of both
types of fluctuations.
Argentina, in particular, had an extremely hard time striking a stable balance in
its development strategy, with political and institutional conflicts undoubtedly
contributing to economic instability. This obviously made it just that much more
difficult to fully implement policies designed to create a more nurturing environ-
ment for the innovation and structural change needed in order to close the gap
with the leaders of the world economy. Brazil marks a sharp contrast in this
respect, since even in the presence of an authoritarian political regime and stark
social inequality, the dominant elites were clearly in agreement on a national
development strategy based on assertive production policies and initiatives for
promoting scientific and technological advances-a strategy that allowed the
country to build up expertise and engage in a process of profound structural
change.
It is hard to know what would have happened if the region had not been hit by a
massive external shock in the early 1980s. It could be that the way in which the
debt crisis diverted the Latin American development process blocked its transition
to a more balanced type of industrialization in terms of the roles of internal and
external markets, of the sort that was starting to take shape in the more successful
East Asian economies.
But there were also other major changes that should not be overlooked. In
particular, sweeping changes were starting to take place in the industrialized world
in the 1970s that took on critical proportions in many countries. As many of the
dynamic elements of the "golden age" of capitalism began to fade and as new
technological paradigms began to emerge, major economic and social changes
started to take place that helped to generate a second wave of globalization. The
Latin American countries were thus faced not only with the difficulties associated
with the declining dynamism (to differing degrees) of what was becoming an
outdated industrialization model and with a looming debt crisis, but also with the
challenges posed by new technological paradigms. While much of the learning
process that had taken place in the countries of the region had consisted of the
appropriation of "Fordist" technologies, the new information technologies that
began to emerge posed new challenges, as did, at the same time, the success of the
industrializing nations of East Asia.
We can therefore say that toward the end of the period being considered here
(and, in the case of the Southern Cone, earlier on), Latin America had narrowed
the gap as much as it could, given its weakness in terms of the development of new
State-led Industrialization 191
technologies. What the region needed to do was not only to open up to the world
economy and reduce state intervention; it also needed to make a qualitative leap in
terms of technological development by making significant advances in developing
national innovation systems in tandem with a parallel effort to improve its
education systems. In the end, it failed to do so, and the debt crisis limited the
policy space to do so. And this is why, as we will see in the following chapter, the
changeover to a more liberal development strategy did not help the region to
overcome the constraints that the industrialization process faced.
Social development
Although some progress had been made during the preceding era of commodity-
export-led growth, major advances in the expansion of the basic education
system's coverage, the development of modern health -care systems and, to a lesser
extent, of social security systems were one of the major outcomes of state-led
industrialization. As was discussed in Chapter 1, this was the period when social
development indicators improved more rapidly than at any other time in the
region's history. As can be seen from Figure 1.7, the greatest improvement in
human development indices relative to industrialized countries occurred between
1930 and 1960, while little or no progress was made in the last two decades of the
twentieth century. The study conducted by Astorga, Berges, and Fitzgerald (2005),
which looks at a somewhat different set of indicators (using the literacy rate as an
indicator of the development of the education system), arrives at the same
conclusion. In addition, as also noted in Chapter 1, the human developll}ent
indices of the different countries of the region converged considerably during
this period, even though sharp differences in per capita GDP remained (see
Table A.4).
Despite the constant fears expressed in this respect, job creation was reasonably
strong. The non-farm workforce grew at the very rapid annual rate of 4 percent in
1950-80, even faster than the rate registered in the United States in 1870-1910.
Although this was reflected in an expansion of informal employment in the cities,
the decrease in employment in traditional agricultural sectors was even steeper.
This led to a decline in total (urban and rural) underemployment in the region as a
whole, which fell from 46 percent in 1950 to 38 percent in 1980. Meanwhile, urban
employment climbed from 44 percent to 67 percent.
Garda and Tokman (1984) identify three patterns in labor trends in the various
countries of the region (see Table 4.12).31 In the Southern Cone, urban employ-
ment predominated and total underemployment was far below the regional
average since 1950, though it increased until 1980 in Argentina and Urugu~y,
owing to the sluggishness of these economies. A second group of countnes
displayed a steep increase in urban employment and a sharp reduction in total
underemployment during this period. This group includes the two largest
economies, as well as the two top performers among the Andean nations
31 See also the earlier study of PREALC (1981), which these authors used as a basis and which
classifies the countries of the region somewhat differently.
192 Economic Development of Latin America
Table 4.12. Relative importance of urban employment and overall underemployment
(% of total employment in each country)
Non- Urban Traditional Overall
agricultural informal agriculture underemployment•
1950 1980 1950 1980 1950 1980 1950 1980
Group A
Argentina 72.0 84.9 15.2 21.4 7.6 6.8 22.8 28.2
8.9 7.4 31.0 29.1
Chile 62.9 77.2 22.1 21.7 4.7 8.0 19.2 27.0
Uruguay 77.8 82.3 14.5 19.0 44.0 18.4 56.9 40.4
47.0 22.0 58.8 36.8
Group B 20.4 9.8 32.7 25.1
22.5 12.6 38.9 31.1
Mexico 34.5 61.5 12.9 22.0 37.6 18.9 48.3 35.4
33.0 18.7 48.3 41.0
Panama 46.7 66.4 11.8 14.8
48.7 37.8 62.7 56.7
Costa Rica 42.0 69.5 12.3 15.3 39.0 33.4 50.7 62.0
39.4 31.8 56.3 51.6
Venezuela 51.1 79.4 16.4 18.5 53.7 50.9 68.7 74.1
35.0 30.1 48.7 49.0
Brazil 39.2 68.1 10.7 16.5 32.6 18.9 46.1 38.3
Colombia 39.2 64.9 15.3 22.3
Group C
Guatemala 30.6 42.7 14.0 18.9
Ecuador 33.3 54.2 11.7 28.6
Peru 36.0 57.5 16.9 19.8
Bolivia 24.1 41.1 15.0 23.2
El Salvador 32.2 47.5 13.7 18.9
Latin America 44.1 67.1 13.5 19.4
*Overall underemployment includes the urban informal sector and traditional agriculture.
Source: Garcia and Tokman (1984}
(Colombia and Venezuela) and the two most developed Central American ones
(Costa Rica and Panama). The members of this group are, for the most part, the
same ones that made the most progress during the period of state-led industri-
alization. The third group comprises the rest of the Andean and Central American
~ountries, which made less progress in terms of urban employment; three of them,
m fact, experienced an increase in total informality.
The large-scale fl?w of rural-urban migration that took place during this period
reflects the generatwn of a labor surplus which W. Arthur Lewis described as
"unlimited supplies of labor." The exception was, of course, the Southern Cone
countries, which had already achieved high levels of urbanization and formal
employment.
The existence of this surplus labor also had important implications in terms of
in~ernational ~igration. Although a few countries continued to attract European
~mgrant~ (parttcularly Venezuela during its long-lasting oil boom), the traditional
mternatwnal flows of migrants began to dry up after the Second World War.
A long-term downtrend in the percentage of Latin American residents who were
born outside the region began in the 1960s as older immigrants died and others
returned t.o their na~ive countr~es. At the same time, intraregional migration
began .to mcrea.se, w1th Argentma and Venezuela being the main destination
countnes, espee1ally for migrants from neighboring nations. And in what was to
pro.ve to be an even more significant trend in the long run, people began to
em1grate to industrialized countries. Between 1970 and 1980, the total number of
State-led Industrialization 193
registered Latin American and Caribbean emigrants to the United States swelled
from 600,000 to 3.8 million,32 and the actual number was no doubt far higher.
Geographic proximity was certainly a major factor in the scale of these flows.
The available data on poverty reduction and, especially, income distribution are
ambiguous and, unfortunately, incomplete. But the records do indicate, at least,
that poverty declined in most of the countries during the period of state-led
industrialization. The earliest available ECLAC estimates indicate that 40 percent
of all Latin American households were poor in 1970, with this figure dropping to
35 percent by 1980 (equivalent to about 40 percent of the population, given the
larger size of poor households). This rate was not to be seen again until a quarter
of a century later, in the mid-2000s (see Chapter 5). The calculations done by
Londono and Szekely (2000: table 2) paint an even brighter picture in the 1970s,
with a reduction in the poverty rate from 43.6 percent in 1970 to 23.7 percent in
1982 and a decrease in extreme poverty from 19.2 percent in 1970 to a low of 10.2
percent in 1981. These calculations may, however, overestimate the positive
trends experienced during those years. Using indirect inferences, Prados de la
Escosura (2007) has calculated that the bulk of the reduction in poverty achieved
in the twentieth century occurred between 1950 and 1980.33
Income distribution continued to be very unequal in most cases but displayed
opposing trends in different countries.34 In sharp contrast with the Southern Cone
countries' poor economic performance, income distribution in those nations
improved, partly as a result of the continuation of a trend that dated farther
back. In the case of Uruguay, inequality indicators pointed to a slight improve-
ment starting in the 1920s. This was associated with the deterioration in the terms
of trade following the First World War and, to a lesser extent, with social and
political changes. The greatest improvement, however, was experienced between
1944 and the mid-1950s, when the sharp increase in the terms of trade, which
would otherwise presumably have produced greater inequality, was counterbal-
anced by industrialization policies, the establishment of wage councils to regulate
recruitment in the private sector, and the expansion of the public sector (Bertola
2005). Argentina had many features similar to those of Uruguay, although the
events of the 1920s suggest that the decrease in inequality during that decade, if it
happened at all, was less marked. As in the case of Uruguay, there is evidence that
inequality decreased significantly as the industrialization process moved forward.
Inequality among wage earners declined (Bertola 2005), and a more general trend
32 Estimated on the basis ofECLAC (2006b) for the Latin American countries, excluding Haiti.
33 According to this author, in six countries (Argentina, Brazil, Chile, Colombia, Uruguay, and
Mexico) poverty levels decreased from 71 percent in 1913 to 27 percent in 1990, with 30 percentage
points of this reduction (i.e., slightly more than two thirds) being registered between 1950 and 1980. See
also the estimates for individual countries in Altimir (2001).
34 The following observations are based on what is a still incomplete body ofliterature, including the
comparative studies conducted by Altimir (1996 and 1997), Frankema (2009), Londono and Szekely
(2000), and Szekely and Montes (2006), whose findings are not necessarily consistent with one another,
and on a number of country studies, some of which are cited here. The statistics compiled by the World
Institute for Development Economic Research (WIDER) of the United Nations University provide the
clearest picture of distributional trends, but the quality and comparability of the data, especially the
figures for years before 1980, are debatable. For an overview of these trends from 1950 on, see
Frankema (2009: table 1.1).
194 Economic Development ofLatin America
can be discerned from the ratio between wages and per capita GDP, particularly
between 1929 and 1950; levels of inequality then remained stable until the advent
of the economic liberalization process and the dictatorship in the late 1970s
(Prados de Ia Escosura 2007). In Chile, inequality continued to increase up to
the 1920s (Rodriguez Weber 2009). This trend was not reversed until much later
on when the emergence of a strong trade union movement, along with the
breakdown of the first wave of globalization, led to social change. It should be
remembered that, historically, the level of inequality in Chile appears to have
always been greater than it was in Argentina.
The lack of comparable data makes it impossible to determine the precise point
in time at which these trends were reversed, except in the case of Uruguay, for
which the data indicate that the favorable trend was interrupted in the mid-1950s
and that it thereafter remained fairly steady until the 1973 coup d'etat, when the
degree of inequality increased drastically. In Argentina and Chile, the reversal of
this favorable trend came later, perhaps not until the early 1970s, but it occurred
under similar political and economic circumstances: bloody military dictatorships
that undercut the traditionally stalwart trade union movements in those countries
and shunned the industrialization strategies that had been in place. In Cuba (the
other country that achieved a substantial degree of economic development early
on), which had a legacy of slavery, income distribution was highly unequal until
the outbreak of the revolution, which brought about the most radical distribu-
tional change in the history of Latin America.
In countries other than those of the Southern Cone, large labor surpluses in the
countryside kept rural wages down, and the influx of rural workers into the cities
also depressed wages for low-skilled urban workers. At the same time, the
shortage of skilled labor, which was caused, among other factors, by the under-
development of most of the countries' education systems, drove up wages for
skilled workers. The fact that labor unions were concentrated in the more formal
sectors of the economy (the government, industrial sectors, and the more modern
services) tended to heighten existing inequalities, although it also broadened the
range of persons who were benefiting from industrialization. This process was
therefore biased in favor of formal urban workers, capitalists, and large rural
landowners.
As surplus manpower came to be absorbed in the cities and the skilled labor
force began to grow, these adverse pressures on income distribution began to ease.
As a result, a number of countries began to see an improvement in income
distribution in the mid-1960s (Costa Rica and Mexico) or early 1970s (Colombia
and Venezuela). This was not an across-the-board phenomenon, however, as
demonstrated by the adverse distributional trend in Brazil during the final
phase of state-led industrialization. In Brazil, as in the Southern Cone countries,
the policies implemented by military governments also had adverse distributional
impacts. This shows that, in addition to the factors associated with the relative
availability of skilled labor, the political regime and its relationship with labor
unions also played an important role in changes in income distribution. These
institutional factors, in particular, had a positive effect in the Southern Cone for a
number of decades, after which they began to have a negative impact (Frankema
2009). The predominance of favorable distributional trends in a larger group of
countries was reflected in an improvement in income distribution in Latin
State-led Industrialization 195
America as a whole in the 1970s (Londono and Szekely 2000), but, unfortunately,
it would not last long.
Overall, the benefits of development were primarily channeled to high-income
sectors, but they also reached the portion of the middle class that was employed by
the government or by large and medium-sized private companies, as well as some
small business persons. The size of the "middle class" varied a great deal, however,
from one country to another, depending on the level of development that each one
had attained by the time that the industrialization period was drawing to a close.
In some cases, this class was very small indeed. This intermediate sector of society
was the first to gain from the emergence of the welfare state and the benefits that it
provided, which were primarily directed toward formal-sector workers (especially
in the case of social security pensions and health-care coverage, as well as the few
unemployment benefit schemes in place). The development of a "Bismarckian
model," under which access to social security coverage was tied to formal-sector
employment, was a decisive factor in this respect. Urban workers that did not
receive these benefits, as well as their rural counterparts, did, however, benefit
from efforts to expand social services (especially education and health care) across
the board. The result was the formation of a segmented (Ocampo 2004a: essay 3)
or truncated (Ferreira and Robelino 2011) welfare state.
The pioneering social security schemes introduced in the Southern Cone
countries during the first wave of such efforts, which dated back to an earlier
phase of development, were given new impetus by the publication of the Beveridge
report in 1942 in Great Britain and by other social security systems that were
being developed in the industrialized world. Brazil had also developed such a
scheme early on, but access to it was much more limited. As a result of this
renewed momentum, a number of countries (Colombia, Costa Rica, Mexico, Peru,
and Venezuela) experimented with what could be described as a second wave of
social security reforms in the 1940s and 1950s. The most noteworthy effort of this
sort was the one made by Costa Rica, which began to introduce reforms in the
1940s that resulted in a system comparable to those of the more developed and
urbanized countries of the Southern Cone. In the rest of the countries of the
region, social security systems would be established later and would be weaker.35
In every case, the system was segmented in terms of the benefits it provided, as
well as being institutionally fragmented, as it was made up of many different
schemes and funds within the framework of a stratified model in which preference
was clearly given to members of the armed forces, civil servants, and strongly
unionized labor groups (Mesa-Lago 1978).
One of the most well-known typologies of welfare states identifies three basic
social security models: a social democratic model, in which the state provides what
tend to be universal, homogenous, non-market-based benefits that are financed by
fairly heavy taxation; a corporatist model, in which the differing levels of power of
different social sectors are reflected in differing benefits and in which access is
afforded by a head of household with a stable job in the formal sector; and a liberal
35 The classic work on the introduction of social security systems in Latin America is that of Mesa-
Lago (1978). The more recent study by Haggard and Kaufman (2008) provides an interesting compara-
tive analysis of the origins and development of the welfare state in Latin America (see, in particular,
ch. 2).
196 Economic Development of Latin America
model, in which the focus is on the poorer sectors of society, while the rest of the
population is left to face social risks within a market-based framework (Esping-
Andersen 1990). The corporatist model was the predominant one in Latin Amer-
ica but, as we have seen, in a highly segmented form in which access was not
provided to informal workers, who made up a very large part of the workforce in
most countries. In Costa Rica, Chile, and Uruguay, as time went by the system
began to evolve in the direction of the first type of model, but only to a partial
extent, since it remained highly segmented. Cuba adopted a social democratic
model after its revolution, as part of its socialist economic model. In Chile, the
reforms undertaken by the military dictatorship eliminated some of these features
and introduced some elements of a liberal scheme. And where the coverage of
these systems was more limited (i.e., in countries other than Costa Rica, Cuba, and
those of the Southern Cone), they were either dual systems or frankly exclusionary
ones (Filgueira 1997).
In rural areas, the benefits of modernization were concentrated in the hands
of large landowners, thus perpetuating those areas' deeply rooted and highly
unequal distribution of income and wealth. There were, however, a number of
agrarian reform programs, with the most ambitious ones being associated with
sweeping social and political change: Mexico in the 1930s, Bolivia and Guate-
mala's failed attempt in the 1950s, Cuba in the early 1960s, Chile and Peru in
the 1960s and early 1970s, and Nicaragua in the 1980s.36 The ground gained by
the last three of these initiatives was lost later on, however. Other efforts, many
of which were spurred by the Alliance for Progress of the 1960s, were less
ambitious. Nonetheless, with the notable exception of Cuba, even the most
radical reform programs did not reach the majority of the campesino population
and failed to do away with the structural dualism characteristic of Latin
American agriculture. The earliest reform program-that of Mexico-left out
half of the campesino population and 57 percent of the land (de Janvry 1981:
ch. 4). The next oldest, that of Bolivia, left out 61 percent of the country's
campesinos and 82 percent of the land; this was largely because it was not
implemented in the eastern part of the country, which went on to develop a
strong agrarian business sector that is now the focus of a heated dispute because
land ownership is highly concentrated. Even at their height, agrarian reforms in
Chile and Peru reached only 20 percent and 32 percent of the campesinos. In the
other countries, they barely scratched the surface of the land ownership struc-
ture, which remained highly unequal (Frankema 2009: ch. 3).
Some small-scale rural producers benefited, in any case, from the opportun-
ity to supply food to the cities and, in countries with open agrarian frontiers,
to settle new areas; this kind of settlement process was promoted in some
countries as part of the agrarian reform program, but, was, in reality, a
substitute for genuine reform (Brazil, Colombia, and Venezuela are the main
examples). Most importantly, however, many campesinos benefited by moving
to the cities where, despite low wages and sometimes inhuman living condi-
tions in the poorest districts, they nonetheless were able to gain a measure of
independence which they had been denied by the coercive forms of labor
36 For a detailed analysis of the history of agrarian reform in the region, see Chonchol (1994).
State-led Industrialization 197
recruitment typical of many rural areas of Latin America in the past. In some
cases, they also gained access to better education, health services, and/or safe
drinking water supplies.
The gradual disappearance of the most servile types oflabor and social relations
that had been typical in rural areas was one of the chief outcomes of the
urbanization process. Agrarian reform programs, even in the countries where
they were the least ambitious, also helped to bring this about. From a long-run
perspective, the subsidence and, in many cases, eventual disappearance of these
servile types of relationships should be regarded as one of the region's most stellar
achievements in terms of equity during this stage of its development. As we have
seen, however, these inroads were made in the midst of the perpetuation of the
stark inequalities inherited from the past and, despite some progress, persistently
high poverty levels in most countries.
5
Turning Back to the Market
The Latin American debt crisis of the 1980s put an end to a period of more than a
century during which Latin America had been gaining ground in the world
economy. It also marked a radical departure from the economic policy that had
been in place since the 1930s. The region's rapid industrialization process had
been slowing down since the mid-1970s, but the debt crisis deepened that trend.
The partial efforts to open up markets that had been made at that time in a few
economies gave way to a widespread move to do so starting in the mid-1980s.
Reforms aimed at expanding the role of the market then began to take firm hold
and remained in place until the first decade of the twenty-first century, when
many countries began to take different economic policy paths in response to
political movements that placed more emphasis on political and social issues and
sought to enhance the role of the state.
The period that we will be analyzing in this chapter encompasses two different
phases. The first, which occurred during the 1980s, was a period of recession
which was aptly dubbed by ECLAC as the "lost decade." The second was a phase
of growth, but was marked by the sharp fluctuations associated with two add-
itional crises: one at the end of the twentieth century, which hit other emerging
economies as well, and the serious worldwide recession of 2008-9, whose epicen-
ter was in the United States. The 3.3 percent annual growth rate registered for
1990-2010 (which is a far cry from the 5.5 percent rate recorded for the thirty-
five-year period prior to the debt crisis) was consequently accompanied by a very
pronounced business cycle. At the same time, the sharp slowdown in population
growth that had begun before this period proved to be a long-term trend. The total
population expanded at an average annual rate of 1.7 percent between 1980 and
2010, one percentage point less than in 1950-80, but that rate also declined
throughout those years and was approaching 1 percent by the end of the period.
This contributed to a somewhat higher per capita growth rate.
This chapter will focus on the changes that have occurred in the Latin Ameri-
can economies since the debt crisis broke out in the 1980s. Given the close
relationship between the macroeconomic events of the late 1970s and the crisis,
the analysis of the first years of the lost decade will tie in with the discussion
presented in the previous chapter. After looking at the origins and implications of
the debt crisis, we will turn to the market reforms that were launched around that
time and their main achievement: the region's more dynamic integration into the
world economy. The chapter will close with an examination of the outcomes in
terms of economic growth and social development. The backdrop for all of these
processes is provided by the major changes experienced in the world economy
Turning Back to the Market 199
during the second wave of globalization, some signs of which had begun to emerge
as early as the 1960s. Those changes included access to a highly volatile inter-
national financial market that has sparked frequent crises throughout the world,
the robust growth of international trade, and, to a lesser extent, the increasing
scale of international labor migration.
THE DEBT CRISIS AND THE "LOST DECADE"
The state-led industrialization model began to come under criticism in the 1960s
both from the advocates of orthodox economic thinking and from the political
left.1 The former criticized it for its lack of macroeconomic discipline and ineffi-
ciencies generated by high tariffs and quantitative import restrictions and, more
generally, by excessive state intervention. The latter criticized it because of its
inability to overcome the economy's external dependence and, in particular, the
unequal social structures that were a legacy of the past. Although he did not
necessarily share the views of the political left, Hirschman (1971: 123) expressed
this idea brilliantly: "Industrialization was expected to change the social order and
all it did was to supply manufactures."
As this model matured, it came to be associated with a great deal of economic,
social, and political tensions. Social conflicts first arose in the Southern Cone
countries, which had been the first to witness major social changes as well as a
slowdown in growth. The opportunities that growth periods provided for enhan-
cing well-being and social entitlements were taken advantage of by social and
political movements, some of socialist inclinations, others with more populist
leanings. However, as the economies were hit by external shocks and balance-of-
payments crises followed, these adjustments fueled growing discontent and op-
position on the part of either popular sectors that demanded greater improve-
ments in social conditions, or the elites, which saw their profit margins threatened
by increased government regulation. The authoritarian response was not long in
coming.
Fishlow (1988: 118) provided a very cogent description of the connection
between social conflict and the transition to market economies in the midst of
the wave of authoritarian military regimes that engulfed the Southern Cone:
"Military instincts are interventionist. But military leaders can conveniently
rationalize political repression in the name of the needed price and wage flexibil-
ity. The objective is not adaptation to a given economic structure but radical
reconstruction of civil society." This implies that the changeover to a market
economy was initially a defensive strategy in reaction to what was seen as an
expansion of socialism. In this sense, the Latin American response differs from
that of the industrialized countries, starting with the election of Margaret
Thatcher in Great Britain in 1979 and Ronald Reagan in the United States in
1980, which were clearly on the offensive. Indeed, the industrialized countries'
1 See, for example, the reviews of this debate by Hirschman (1971), Fishlow (1988), and Love
(1994).
200 Economic Development of Latin America
response was a reflection of the confidence of private enterprise that it could do
without state protection and even the belief of large sectors of the business
community that state intervention had actually become an obstacle to its devel-
opment. This offensive would be taken up by Latin America later on, chiefly in the
last decade of the twentieth century.
Outside of the Southern Cone, although social unrest was also on the rise, it did
not have any direct link with the transition to market economies. In Central
America, which became the epicenter of the conflict in the 1980s, confrontations
occurred primarily in rural settings and stemmed from the concentration of land
ownership and, perhaps, from the commodity-export-led growth model rather
than from its somewhat peculiar connection with a feeble form of state-led
industrialization. Colombia's long history of internal conflict was also rooted in
rural problems, but starting in the mid-1980s, it took on quite a different character
as drug trafficking gained ground, which provided funding for all sorts of violence:
rural conflicts, paramilitary groups, and guerrillas (or, at least, funding for the
largest guerilla organization). The violence associated with drug trafficking would
later engulf Mexico and Central America in the first decade of the twenty-first
century.
As discussed in the preceding chapter, the lack of macroeconomic discipline
was less widespread than it is often portrayed as having been. In fact, it was
primarily a problem in Brazil and the Southern Cone, rather than in the rest of the
region, at least until the mid-1970s. However, the tendency to run an external
deficit, which had indeed been a long-standing trend, was growing stronger
toward the end of the phase of state-led industrialization in almost all the
countries of the region. This was the result of both the behavior of the trade
balance and an increasing demand for investment (which economic theory tells us
are actually two facets of the same problem). These imbalances were dealt with by
resorting to greater and greater amounts of external borrowing. This, however,
turned out to be the sword of Damocles of the model, given the volatility
associated with such financing. Figure 5.1 depicts the first of these trends. As the
reader will see, until the industrialization process was quite far along, growth was
coupled with small trade surpluses (generated, as seen in the preceding chapter, by
massive balance-of-payments interventions). It can even be said that the small
deficit registered in 1967-74 was not a problem, given the striking upswing in
growth experienced during those years. It proved impossible, however, to sustain
growth during the period between 1974 and 1980 (at rates not unlike those
observed prior to 1967) without generating a deepening trade deficit.
Growth was also associated with increasing investment requirements that
countries with endemically weak national savings rates found hard to meet. The
investment rate had fluctuated between 19 percent and 22 percent of GDP up to
the mid-1970s, reaching its lowest point in 1958-67, the years of what ECLAC
called "external strangulation" (see Chapter 4). It climbed to 25 percent during the
final phase of this stage of development (see Table 5.1). This indicates that the
higher levels of external borrowing seen during the 1970s were reflected in higher
investment rates (which no doubt included a number of white elephants in some
countries), in sharp contrast to subsequent periods, when higher levels of external
borrowing instead drove up consumption.
Turning Back to the Market 201
-1.5% -1.0% -0.5% 0.0% 0.5% 1.0% 1.5% 2.0%
Trade balance (% of GOP}
Figure 5.1. Economic growth and trade balance
Source: Authors' estimations based on ECLAC historical series
Table 5.1. Gross fixed capital formation (% of GDP) 1991- 1998- 2004- 2008-
1997 2003 2008 2010
1950- 1958- 1968- 1975- 1981-
1957 1967 1974 1980 1990
Simple average 23.9% 20.1% 21.6% 24.3% 19.1% 19.6% 18.3% 21.5% 23.3%
Large countries 14.2% 15.7% 18.1% 21.5% 17.0% 19.2% 20.0% 19.8% 19.1%
Small countries 19.1% 17.6% 19.5% 22.6% 17.8% 19.4% 19.4% 20.5% 20.8%
Latin America
19.5% 22.2% 25.1% 18.9% 18.2% 18.0% 19.9% 20.9%
Weighted average 16.8% 17.7% 22.2% 16.9% 18.6% 19.3% 19.1% 18.7%
Large countries 21.0% 19.1% 21.9% 24.9% 18.8% 18.2% '18.1% 19.8% 20.7%
Small countries 15.8%
Latin America 20.7%
Source: ECLAC historical series at constant prices. Costa Rica data are available since 1952, El Salvador since 1962,
Nicaragua since 1960, and Uruguay since 1955. The figures correspond to the average of all countries for which data
are available.
State-led industrialization also ran up against other constraints: those associ-
ated with the tendency to overwhelm the state with fiscal responsibilities without
giving it sufficient resources to meet them. As noted by FitzGerald (1978), this was
reflected in three main trends: (i) an upward trend in public expenditure as a
proportion of GDP, combined with a downward trend in the share of spending on
social welfare programs relative to industrialized countries; (ii) a shift in the
composition of the tax structure away from property and income taxes and
toward indirect taxes and taxes on wage income; and (iii), as a result, rising
borrowing requirements, given the need to finance transfers to the private sector
rather than redistributive social policies. This was particularly a problem in the
second half of the 1970s, when the widespread access of Latin American countries
202 Economic Development of Latin America
to external financing led to rising fiscal deficits, which then made public-sector
accounts highly vulnerable to any tightening of external credit, which eventually
did, in fact, occur.
It is unlikely, however, that if the debt crisis had not occurred, any of the Latin
American economies would have collapsed under the weight of the inefficiencies
generated by state-led industrialization or of these types of macroeconomic
tensions. What is more, it is unclear why they could not have adopted or further
developed a more balanced strategy, as the smaller countries had begun to do in
the mid-1950s and most of the mid-sized and larger countries began to do in the
mid-1960s. As was discussed in the preceding chapter, early on the region had
begun to take advantage of the opportunities which the growth in world trade had
started to open up and had evolved toward a mix of protection and export
promotion. In fact, the literature of the 1970s portrays a number of Latin
American countries, particularly Brazil, as international export success stories
on a par with the Asian tigers.
Thus, the Latin American countries could have moved toward a development
model more along the lines of the East Asian model, which was also state-led and
also somewhat protectionist, but which placed more emphasis on building a solid
export base and, in most cases, showed a clear preference for national over foreign
investment. As it turned out, however, the scale and speed of events ruled that
option out. We have also noted in the previous chapter that this was not the only
possible path. Earlier on, the Southern Cone countries had in fact taken a different
route, with slow growth combined with an improvement in distribution and
mounting social conflict.
These long-term trends notwithstanding, what sounded the death knell for that
paradigm was the boom-and-bust cycle of private external financing, which began
slowly in some countries in the mid-1960s, spread out to the rest of the region in
the 1970s, and culminated in the debt crisis of the 1980s. This kind of cycle had
been experienced before, most recently in the boom-bust cycle of external finan-
cing of the 1920s and early 1930s. The sources of external finance were different,
however, as syndicated credits from theinternational commercial banking system
now took over the role that bonds floated on international capital markets had
played in the 1920s.
One of the conspicuous features of the quarter-century following the Second
World War had been the absence of large volumes of private external financing
and the rather modest level of official finance. As shown in Figure 5.2, net resource
transfers from abroad2 were slightly negative in the 1950s and 1960s. Against the
backdrop of recurrent external shocks, the fact that countries lacked sufficient
means to cover their balance-of-payments deficits, including the very modest
financing available from the International Monetary Fund (IMF), obviously
heightened the temptation to resort to protectionist policies as an adjustment
mechanism. The countries that were the first ones to gain access to private
external financing (Mexico and Peru, in particular) were also some of the first
to run into problems of over-indebtedness.
2
Net resource transfers are defined as the balance on the capital account minus debt service
(interest payments on the external debt and dividends sent abroad by foreign corporations).
Turning Back to the Market 203
4%
2%
0%
-2%
-4%
-6%
-8%
0 ll) 0
ll) ll) CD
0> 0> 0>
I Io % of GOP 1• Via FDI • Via financial flows
Figure 5.2. Net resource transfer(% of GDP at current prices)
Source: ECLAC historical series
The new boom in external financing for Latin America was part of a broader
move to rebuild the international capital market that had first taken shape in the
1960s (when it was dubbed the "eurodollar market"). The hallmark of this process
was competition among a growing number of formerly national banks that had
begun to operate as international institutions that provided financing in global
markets, generally syndicated loans at variable interest rates pegged to the three-
or six-month London Interbank Offer Rate (LIBOR). This mode of operation
facilitated the entry of smaller banks with less international experience, which
trusted almost blindly in the credit evaluations of the large banks that led the
process (and that received hefty commissions). By pegging the interest rate to the
interbank market, which was the source of financing for banks actively involved in
the international market, the risk for creditors represented by variations in those
rates was reduced by shifting it onto borrowers. As we will see, these risks became
dramatically evident since late 1979 and ultimately proved to be disastrous. These
laxly regulated banks first ran into problems in late 1974 when some of them,
particularly the Herstatt Bank in Western Germany and the Franklin National
Bank in the United States, lost heavily on foreign exchange operations. The
recycling of petrodollars on that market in the following years gave it a strong
boost that was reflected in the abundant financing received by the region in the
second half of the 1970s (Devlin 1989: ch. 2).
Within an oligopolistic setting, in which large banks sought to place loans in a
way that would allow them to expand or at least maintain their market share,
external lending activity began to increase steeply and was leveraged by the
additional resources provided by smaller banks, generally at small spreads over
LIBOR (between one and two percentage points, with the spread usually being
closer to one point as the boom neared its end). High levels of liquidity in the
eurodollars market and low real interest rates (which at some points were actually
negative) in the 1970s combined with high commodity prices (for oil, in
204 Economic Development of Latin America
particular, but for other products as well) to generate strong incentives for heavy
external borrowing (Devlin 1989; Ffrench-Davis, Munoz, and Palma 1998}. In
fact, Latin America accounted for over half of all private debt flows to the
developing world in 1973-81 (Ocampo and Martin 2004: ch. 3) while at the
same time continuing to be the developing region that attracted the largest
share of foreign direct investment (FDI). ·
The counterpart of booming lending was the growing trade and fiscal deficits
that the region built up. National financial institutions that served as intermedi-
aries for transactions involving those external funds also began to find themselves
taking on higher and higher levels of credit and foreign exchange risk. This problem
was, however, associated with a new trend: liberalization of domestic financial
markets. This is why it was more serious in the countries of the Southern Cone,
since they were the first to undertake market reforms. The governments' ability to
enforce exchange controls aimed at preventing capital flight once the crisis had
broken out was also an important factor. Capital flight occurred throughout the
region, but took place on a massive scale in Argentina, Mexico, and Venezuela,
which lacked sturdy mechanisms for controlling capital movements.
The differing sizes of the various countries' external and fiscal deficits and the
differing degrees of their financial systems' fragility played a crucial role in
determining the relative impact of the 1980s debt crisis. This indicates that the
countries' macroeconomic dynamics, rather than defects in the production struc-
ture created by the preceding model, were the decisive factor. And this is why the
problem arose both in the more tightly regulated economies (e.g., Brazil) and in
the more liberalized ones (those of the Southern Cone). Indeed, in financial terms
the problem was most serious in the latter countries, where it triggered some of
the most dramatic domestic financial crises in history. Moreover, the fact that
Latin American exporting countries had faced similar difficulties in striving to
manage the sharp external financing cycle of the 1920s and 1930s and that the
more liberalized economies were confronted with a similar situation in the 1990s
(see below} indicates that boom-bust cycles fueled by the volatility of external
financing is a general phenomenon rather than a feature of state-led industrializa-
tion as such.
This is why external shocks played such a pivotal role in determining how the
crisis unfolded (ECLAC 1996: ch. 1). The turning point was the decision, made in
late 1979, by the Federal Reserve Board of the United States to raise interest rates
steeply (this became known as the "Volker shock," after the Federal Reserve
Chairman of the time) in order to stamp out the inflationary spiral that the US
was experiencing at the time. This had a direct impact on the debt service, since
much of Latin America's external debt had been contracted at floating interest
rates. This situation was compounded by a sharp drop in the real prices of raw
materials. Both of these adverse shocks were to last nearly a quarter of a century.
This factor, which, of course, is only apparent in hindsight, is generally overlooked
in analyses of this period of economic history (see Figure 5.3}.
Real interest rates in the US had been very low right up to the 1960s and were
actually negative in the mid-1970s, but then shot up in the late 1970s and
remained high for the rest of the century, and this was especially true of long-
term rates. This pattern was even more marked for the rates relevant for Latin
America. The real effective interest rate on the Latin American region's debt
Turning Back to the Market 205
'
fluctuated between -1 percent and 2 percent between 1975 and 1980 (estimated at
one percentage point above the three-month LIBOR ~nd with c~rrent inflati~n
rates}. Even taking into account the subsequent rate htkes (what 1s referred to m
Figure 5.3 as the "ex-post rate"),3 it averaged no more than 4 percent during tho~e
years, reaching a peak of 6 percent in 1981-2. In contrast, when the Latm
American countries rt;turned to the capital market in the 1990s (when the
reference rate had become the rate on ten-year US Treasury bonds), the real
interest rate generally stayed above 10 percent (once the corresponding spreads
are factored in). Thus, the region did not again see rates similar to those charged
in 1975-80 until the international financial boom of 2005-8.
The decline in commodity prices also proved to be a long-run break from the
earlier trend and would last until the mid-2000s (Ocampo and Parra 2003 and
2010). At their lowest point, between 1992 and 2001, real commodity prices were
37 percent (and at times as much as 40 percent) below their average for the 1970s,
which was in turn actually slightly below the average for 1945-80. These two long-
run adverse factors were joined, in the early 1980s, by a sudden slowdown in the
industrialized world and an outright recession in the US.
International interest rates had never before been so high for so long.4 Reces-
sions such as those experienced by the industrialized countries had, on the other
hand, occurred before, as had a long-term steep decline in the terms of trade. In
the first case, however, the 1982 economic slowdown in the industrialized world
was somewhat steeper than that of 1975, and was thus the worst of the post-war
period (until it was surpassed by the deep 2008-9 recession}. In the case of the
terms of trade, the last time that anything similar had occurred had been when
commodity prices had plummeted in the 1920s and 1930s. Consequently, the
ex-post risks that Latin America had to assume were not only unexpected, but also
quite difficult to foresee.
The debt crisis erupted after the shock generated by the hike in interest rates.
External debt coefficients had been climbing steadily, but slowly, since the 1970s
and, on average, were still moderate in the 1980s (below 30 percent of GDP on
average and slightly more than two times the value of exports), thanks, n~ doub.t,
to the favorable conditions associated with the boom. In the years followmg thts
period, a steep increase was seen in those coefficients as a result of. s~arply high~r
interest rates, sinking commodity prices, and the even more prectpttous drop m
Latin America's GDP, measured in dollars, which was in turn caused by the
combination of a deep recession with the huge devaluations triggered by acute
foreign exchange shortages. In slightly more than half a decade, Latin America's
external debt coefficients had doubled and as a consequence of the long-run
3 This real ex-post interest rate was calculated as the average annual rate for the y~ar in which t~e
loan was taken out and the six following years (based on the assumption that a loan typiCally matures m
seven years) using the LIBOR + 1 as the nominal rate and the US consumer price index as a deflator.
4 The deflation associated with international crises up until the 1930s did drive up real short-term
interest rates. These increases were strictly temporary (lasting for three years during a serious crisis
such as the Great Depression of the 1930s) however, since as nominal interest rates began to decline as
a result of the crisis, real rates came down rapidly-so much so, in some cases, that they became
negative in real terms.
206 Economic Development of Latin America
A. Real Interest Rates
18.0,----------------------
- - 10-year U.S. Treasury bond rate - - - U.S. 3-month Libor
- Effective rate-Latin America - LA ex-post debt cost
B. Real Non-Oil Commodity Prices (1980=100)
140.0 l~------------------~-----
130.0 1~----~-------------------
120.01~----t-ll-------------------------
11 0.0 ~-~..-------cl
100.0
90.0i~---------\-~\------------L--
80.0 1~----------_____]t-------~---------l-----
70.0
60.0i---------------~~---~-----
50.0 +-r-..-,--,-,.--,-,rr-T--ro--r-r.....-........,..,--..-,....,.....,.--,-,rr-T--r-T-,--,-...,.,..-r-r'-.---r-~~~~
Figure 5.3. A. Real interest rates B. Real non-oil commodity prices (1980 = 100)
Sources: A. Authors' estimations based on Global Financial Data, Inc. for Libor rates; US Federal Reserve for the
Treasury rates; and Data Stream to calculate the effective rate of Latin America. The latter is estimated as Libor + 2 in
1975-85 and the _yield oft~e Latin American b?nds from 1993 on, according to j. P. Morgan. (For the 1993-7 period,
these were re-estimated with the data on the )'leld of the Treasury bonds and Latin American EMBI.) The US CPI is
used as a deflator in all cases.
B. Data updated with the sources listed in Ocampo and Parra (2010).
Turning Back to the Market 207
60%450o/o----~--------------------------------------------T
50%
40%
-cc.
30% (.!)
0
>0!!.
20%
10%
50o/o ~--------------------------------------------------
0% 0%
0r-- Nr-- 'r<--t Cr-D- 0r-0- 0 N '<t CD 00 0 N '<t CD 00 0 N '<t CD 00 0
00 00 00 00 00 a> a> a> a> 0 0 00 0 0
a> a> a> a> a> a> a> a> a> a> 0 0 0 0 0 N
a,.>... a,.>... ,a..>.. a,.>... a,..>.. N N N N N
a,..>..
1-- % of exports - %oiGDPI
Figure 5.4. Dynamics of the Latin American external debt (% of GDP and exports)
Source: Authors' calculations based on data of external debt from the World Bank, and nominal GOP and exports
from ECLAC historical series. The data for 2010 were updated with the growth rate of debt according to the World
Bank.
factors mentioned above, did not drop back to their pre-crisis levels until the first
decade of the twenty-first century (see Figure 5.4).
The situation reached dramatic proportions as the adverse conditions persisted
and the international policy response to the debt crisis in Latin America (and in
some other parts of the developing world) proved to be quite feeble. The com-
bined effect of the sudden and protracted (nearly decade-long) absence of external
financing and mounting debt service generated a massive external shock that
turned the region's positive net resource transfers, which had been equivalent to 2
percent or 3 percent of GDP, into negative transfers amounting to about 6 percent
of GDP (see Figure 5.2).
Diaz-Alejandro (1988: 310) summed up all of these events masterfully when he
said that: "what could have been a serious but manageable recession has turned
into a major development crisis unprecedented since the early 1930s mainly
because of the breakdown of international financial markets and an abrupt change
in conditions and rules for international lending. The non-linear interactions
between this unusual and persistent external shock and risky or faulty domestic
policies led to a crisis of severe depth and length, one that neither shocks nor bad
policy alone could have generated."
The inherent instability of international financing cycles (a feature discussed
in Chapter 1) thus proved to be a decisive factor in determining the fate of
the development model based on primary exports and the era of state-led
industrialization.
208 Economic Development of Latin America
A comparison with the 1930s will help to show just how crucial the effect of the
negative resource transfers of the 1980s was. As illustrated in Figure S.S.A, the
opportunities for boosting exports and their purchasing power were much greater
in the 1980s than they were in the 1930s. Thus, the crucial difference between the
debt crisis and the Great Depression was the massive, long-lasting shock to the
capital account. This situation was not properly addressed at the international
level, with the result that the region sank into the worst crisis of its entire history.
As the prospect of bank failures loomed for overexposed banks worldwide and,
in particular, in the US (Latin America's debt was equivalent to 180 percent of the
capital of the nine largest US banks), the US and other industrialized countries'
governments put pressure on the IMF and multilateral development banks to run
to the rescue and started freeing up larger amounts of credit than they had in the
past. The funds that they made available were, in any case, modest in comparison
to the impact of the large-scale turnaround in private resource transfers and were
also accompanied by unprecedented "structural" conditionalities (which took the
form of what were, in most cases, draconian market reforms and fiscal adjust-
ments). As was seen in the preceding chapter, in the 1930s external debt defaults
proved to be a solution for most of the countries involved, just as they had been in
all the previous external debt crises. As the 1980s unfolded, temporary "silent
defaults" in the form of arrears in the servicing of commercial and bilateral (and,
in a very few cases, multilateral) debts became more and more frequent. This was
partly because of the internal tensions that this overly prolonged crisis began to
generate in a region which was, moreover, witnessing a return to democracy
(Altimir and Devlin 1993). Be this as it may, the strong pressure brought to
bear by industrialized countries and multilateral agencies prevented the Latin
American countries from openly declaring defaults and pushed debtor countries
into concluding renegotiation agreements that were clearly advantageous for the
commercial banks involved. The 1989 Brady Plan opened the way for a few debt
write-offs, but the amounts involved were moderate and the cancellations came
too late to head off the damage caused by the debt crisis.
As a result, while, during the 1930s, the Latin American economies simply had
to increase their trade surpluses for a fairly short period of time, in the 1980s they
had to generate hefty trade surpluses for a period lasting over a decade (see
Figure S.S.B). The combined impact of all of this turned out to be that, while
the initial effect of the Great Depression on per capita GDP in the Latin American
economies was more severe, their subsequent recovery was quite strong and, from
1937 on, per capita GDP was invariably above pre-crisis levels. In contrast, the
recovery from the debt crisis of the 1980s did not come until 1994, i.e., fifteen
years down the road.
Three different stages in the debt crisis can be identified.5 In the period up to
September 1985, large-scale macroeconomic adjustments were made on the
assumption that the crisis would be short-lived (i.e., that it was a liquidity crisis
5 See, among many others, Devlin (1989), Altimir and Devlin (1993), and Ffrench-Davis, Muiio~.
and Palma (1998). Devlin divides each of the first two phases into two subperiods of debt renegoti-
ations. The conditions associated with the various phases of the negotiations are covered in detail in the
first chapter of Devlin (1989) and in the editions of ECLAC's annual Economic Survey ofLatin America
and the Caribbean published during those years.
Turning Back to the Market 209
A. Purchasing Power of Exports
160~------------------------------------------------
140
120
\100 ~
80
~ ____ ~-~60 ............... ~/
40+---.----.---.----,---.----.---.----.---.---.----,
co 0) 0
"'0
C\J (') '<!' LO CD
@ @ @ @ @ @ @ @ @ @ @
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
l-1930s (Year 0=1929) I1980s(Year0=1980)
B. Trade Balance as% of Exports (vs 1929 and 1980)
40%
30% ----------
20%-1------
10%
-10% --------------------
-20"/ol--o_ _ _ _ _ _ _C_\J _ _(_') _ _-_.t ___L_O _~---~-~~--~-~~
CD co 0) 0
"'~
~ ~ ~ ~ ~ @ @ @ @ @
~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~
Il-1930s (Year 0=1929) 1980s (Year 0=1980)
Figure 5.5. Comparison of the crises of the 1930s and 1980s
A. Purchasing power of exports
B. Trade balance as% of exports (vs 1929 and 1980)
Source: Authors' calculations based on ECLAC (1976) for the 1930s and ECLAC historical series for the 1980s
210 Economic Development of Latin America
rather than a solvency crisis) and that voluntary lending would soon make a
comeback.6 There was also a powerful cartel of lenders that had the backing of the
governments of industrialized countries, which intervened because they felt that
their financial systems were under serious threat. On the other hand, although
some governments adopted more radical stances, such as the decision taken by
Alan Garda in 1985 to limit Peru's debt service to 10 percent of its export
earnings, and although some unconvincing attempts were made to form some
sort of association of debtors (the 1984 Cartagena Consensus being the best-
known), there was never an effective move to form a "debtors cartel," which, if it
had actually come into being, would no doubt have trigg~red a severe crisis in the
private international banking system, especially in the US. The measures that were
adopted were, therefore, very effective in averting a financial crisis in the US, but
entirely inappropriate for handling the Latin American debt crisis. What is more,
because of the asymmetrical nature of the negotiations, the Latin American
countries ended up "nationalizing" large portions of the private external debt.
Thus, Latin America can rightly be seen as a victim of the way in which what was
also a US banking crisis was handled; oddly enough, this is not fully recognized in
the existing literature.7 The great irony was that, as a result, US banks were
turning a profit while Latin America slipped into the worst economic crisis of
its history (Devlin 1989).
In September 1985, the crisis entered into a second phase with the announce-
ment of the first Baker Plan, which provided for a structural adjustment headed
up by the World Bank, better lending terms, and a modest amount of fresh credit.
This package was insufficient, however, and, two years later, was replaced with a
second Baker Plan which added debt buybacks, low-interest exit bonds, and debt
swaps. The final phase began in March 1989 (i.e., nearly seven years after the
outbreak of the crisis) with the Brady Plan, which included a modest reduction in
debt balances and was soon followed by renewed access to private external
financing. The United States' involvement in these last two phases differed from
its approach in the first, with the authorities working to offer actual solutions for
what was now clearly seen as a solvency crisis at a time when there were also signs
that Latin American countries were increasingly reluctant to continue following
the earlier approach.
Although the Baker Plans and, especially, the Brady Plan finally led to reduc-
tions in the countries' external debt coefficients (see Figure 5.4), the upward trend
in those coefficients had already been reversed by the large trade and current-
account surpluses that the countries built up, at the cost of a "lost decade" in terms
of economic growth. That loss represented a drop of somewhat more than 8
percent in their per capita GDP. Latin America's share of world GDP, which had
been expanding for over a century, fell by 1.5 percentage points, while its per
capita GDP shrank by eight percentage points relative to that of industrialized
countries and by twenty-three percentage points relative to the world average (see
Table 1.1 of Chapter 1).
6 Cline (1984), who authored what is perhaps the most well-known presentation of this view, argued
that the crisis would be overcome once the industrialized economies made a recovery.
7 In fact, it makes little sense that this crisis is not classified in the relevant databases as the US
banking crisis that it actually was. See, for example, the IMP database (Laeven and Valencia 2008).
Turning Back to the Market 211
The recession was initially very severe. The region's GDP shrank for three years
in a row. The contraction was particularly sharp in 1983, when the full impact of
the Mexican default of August 1982 made itself felt. This is generally considered to
be the starting point for the debt crisis (see Figure 5.9 below). In 1984-7 there was
a moderate recovery, but the situation deteriorated in the closing years of the
decade. Few countries were able to put their economies back onto a stable growth
path in the second half of the 1980s; those that did were generally countries with
moderate external debt coefficients (Colombia) or ones that received fairly hefty
amounts of official external financing (Chile and Costa Rica). As will be seen later
on, the decrease in per capita income was accompanied by a steep reduction in the
manufacturing sector's share of economic activity.
The social costs of the crisis were huge. The poverty rate climbed sharply
between 1980 and 1990 (from 40.5 percent to 48.3 percent of the population).
The deterioration of income distribution in a number of countries exacerbated the
sharp inequality that was already a long-standing feature of Latin America, and
reversed the progress that had been made in this respect during the 1970s by a
number of individual countries and by the region as a whole. This was, in most
cases, accompanied by a decline-a steep one in some cases-in real wages in the
formal sector and the expansion of informal employment. The rapid improve-
ment in human development indices made during the period of state-led indus-
trialization gave way to a much slower rate of progress and to an actual
deterioration in some areas (see Chapter 1, Table A.4, and below).
Huge adjustments in fiscal and monetary variables and in the exchange rate put
added stress on what were already precarious economic structures. The depreci-
ation of the real exchange rate, which was necessary in order to support external-
sector adjustment, was inevitably accompanied by a surge in inflation, which
reached proportions never before seen in Latin America, even taking into account
the inflationary histories of some countries. Inflation had sped up in the 1970s, as
was happening elsewhere in the world as well, and two countries had ushered in
the era of triple-digit inflation in the midst of serious political crises (Chile and
Argentina). Nonetheless, inflationary spirals were an effect rather than a cause of
the debt crisis. The most bizarre manifestation of this effect was the bouts of
hyperinflation that overtook five countries in the mid-1980s and early 1990s
(Argentina, Bolivia, Brazil, Nicaragua, and Peru). Another three registered
triple-digit inflation at some point as well (Mexico, Uruguay, and Venezuela).
Panama (the only dollarized economy at the time) was the only country in which
inflation did not climb above 20 percent. For the region as a whole, as shown in
Figure 5.6, the median and mean rates of inflation soared, reaching nearly 40
percent and over 1,000 percent, respectively, in 1990, before beginning to subside in
the years that followed. The crises that broke out in the financial sector were also
massive, especially in the Southern Cone, where they took a toll in terms of fiscal
and quasi-fiscal costs equivalent to as much as 40 percent or 50 percent of GDP.8
8 See Laeven and Valencia (2008), which make it clear that the financial crises that hit the three
countries of the Southern Cone in the early 1980s were some of the most costly to be seen in the last
three decades and are actually comparable only to those experienced by some East Asian countries
during the 1997 crisis.
212 Economic Development of Latin America
A. Median
40.0,-----------------------------------------------
35.0 t - - - " - - - - - - - - - - - - - - - - - - 1 \~----~---~
30.0 --~--~-"
25.01--------
20.0 · - - - - - - -
15.0 1 - - - - - - - - - - / - - \
1-- ILatin America (18 countries)
B. Mean (annual percentage change in natural logarithm)
1000.0 ,_ _ _ _ _ - - - - - - - - - - - - - - - - - - . 4
100.0 --------------------------~1
1.mg0~m~~~m~-=m~~~m~~~m~--m~~~m~--m~--m~--m~--m~--m~--~m--m~--~m--m~--0o --o~--o~--o~
~ ~~ ~ ~~~ ~~ ~~ NNNN
1-- ILatin America (18 countries)
Figure 5.6. Inflation rate in Latin America (CPI, annual percentage change)
A. Median
B1 Mean (annual percentage change in natural logarithm)
Source: IMP, International Financial Statistics
The distribution problems that arose within the countries as they strove to cope
with the crisis were closely associated with the need to make transfers to the
governments so that they could service their countries' external debt and pay
the costs of the collapse of their domestic financial systems. These transfers could
be made more easily in countries where the state had direct access to hard-
currency export earnings (mainly through state-owned oil and mineral
Turning Back to the Market 213
enterprises) and where the government consequently benefited directly from the
devaluations. Others were confronted with a serious "domestic transfer problem"
as they struggled to find ways of transferring fiscal resources to the state for use in
servicing the public debt; as such service rose in terms of the local currency
because of the devaluations, it became even more difficult to cover (ECLAC
1996; Altimir and Devlin 1993).
The adjustment also entailed an enormous reduction in investment (a drop of
six percentage points from its 1975-80 peak, as shown in Table 5.1), even though
domestic saving was on the rise. In this last case, the domestic transfer problem
made it necessary to reduce the real income of wage-earners (the sector with the
greatest propensity to consume) or, more often, oblige them to undertake "forced
saving" via inflation. Against a backdrop of growing distributional conflict, this
situation was reflected in surging inflation and in the high social costs of the
adjustment.9 Meanwhile, the investment rate would take a quarter of a century
(i.e., until the 2004-8 boom) to regain the levels attained prior to the mid-1970s,
but it has still not returned to the levels achieved in the second half of the 1970s
(see Table 5.1). There is, moreover, a consensus that forcing governments to cut
back on infrastructure investment as part of the adjustment program stunted
long-term growth (Easterly and Serven 2003).
MARKET REFORMS IN THEORY AND IN PRACTICE
A key difference between the new and old paradigms was the relationship between
theory and practice. As was seen in the preceding chapter, in the case of state-led
industrialization, the theory developed by ECLAC came at a time when the
process was already quite far along. ECLAC's tenets thus provided an analytical
foundation and lent greater rationality and coherence to a practice that had been
in place for several decades or, in some cases, even longer. In the new paradigm,
on the other hand, the theory came first in the guise of an intellectual and openly
ideological assault which, although it had precedents, truly came into its own in
the 1970s. The most paradigmatic and earliest case was, of course, the offensive
mounted by the "Chicago Boys" in Chile starting in the 1950s, which held sway
during the Pinochet regime, leaving its imprint on a regime that had initially
lacked any economic model at all (Valdes 1995). A number of writings intended to
spread the new ideas, including that of Balassa et al. (1986), played an important
role in this regard at the regional level.
Starting in the 1980s, explicit institutional support for these new policies was
provided at the international level by the World Bank, which played a central role
in pushing for "structural reforms." It was joined by the IMF, which espoused the
more specific features of macroeconomic adjustment. The industrialized world,
for its part, provided demonstration effects through the ideological and political
schools of thought that led to Thatcher's and Reagan's rise to power.
9 For a discussion of the various dimensions of the domestic transfer problem, including the fiscal
transfers mentioned above, see Frenkel and Rozenwurcel (1990).
214 Economic Development of Latin America
The impact of these external factors marks a contra~t with the transition to the
paradigm of state-led industrialization, which, although influenced by external
schools of thought and policies in the industrialized world, was clearly an
endogenous process. Thus, whereas the text that summed up the earlier approach
most cogently was ECLAC's "Latin American manifesto" (see Chapter 4), the new
paradigm was most clearly set out in John Williamson's (1990) decalogue, in
which he outlined the reform agenda that international financial institutions had
decided that the Latin American countries should adopt (rather than his own
ideas). The "Washington consensus" soon became virtually a synonym for market
reforms. This marked a definitive shift toward the economic thinking that held
sway in the industrial economies, especially the United States.
Nonetheless, the different countries' approaches did vary a great deal, even
during the height of the reform process. Stallings and Peres (2000), who analyzed
the experiences of eight countries of the region (nine, counting Jamaica), divide
them into two groups: the "aggressive" reformers (Argentina, Bolivia, Chile, and
Peru), and the "cautious" ones (Brazil, Costa Rica, Colombia, and Mexico). The
first group introduced sweeping reforms within a very short period of time, while
the second took a number of years to do so and moved forward at differing speeds
in different areas. Most of the countries of Latin America probably fall into the
second group. In fact, in many cases, the idea of "neo-liberal" policies, which has
been used by many authors, is perhaps not the most appropriate description of
the reforms, since they entailed state intervention on a scale that is at odds with
more orthodox economic thought. Accordingly, the more generic term "market
reforms" will be used here.10 The existence of this range of approaches also
indicates that the reform process in the region should not be seen simply as
something that was imposed on the region from outside: it was the outcome of
decisions which, unlike the early experiments made in the Southern Cone (that
can be accurately described as "neo-liberal" in nature), were made within the
countries by democratic governments. In fact, for the first time in Latin America's
history, at least at the regional level, economic liberalism joined hands with
political liberalism.
Thus, reforms aimed at liberalizing the market and reducing the scale of public-
sector activity in the economy were coupled with macroeconomic stabilization
policies designed to correct the countries' external and fiscal deficits and to bring
the inflationary spiral triggered by the debt crisis under control. This combination
of measures has given rise to a great deal of confusion in analyses of the countries'
reform processes. The more aggressive reformers introduced major liberalization
measures in conjunction with macroeconomic stabilization plans (Chile in the
mid-1970s, Bolivia in the mid-1980s, and Argentina and Peru in the early 1990s),
but this was far from being the general pattern. It is important to draw a
distinction between these two types of policies because there is no binding link
between them; in other words, it is quite possible to attain macroeconomic
10 The term "liberal" is used to mean very different things in different countries. In the English-
speaking world, it is nearly a synonym for state intervention, and this is the way in which it has been
used by some liberal schools of thought in Latin America. Consequently, some authors prefer to use the
term "neo-conservative" to refer to these market reforms.
Turning Back to the Market 215
stability in economies that have not been liberalized, and liberalized economies
may still exhibit major macroeconomic disequilibria.
The timing and diversity of the different countries' reform processes can be
discerned from the structural reform indices calculated by Morley et al. (1999),
who divide the reform process into two different phases. The first took place in the
1970s, was gradual, varied a great deal from country to country, and lost ground
during the first phase of the debt crisis. The second phase, which has also been
tracked by Lora (2001), was, on the other hand, a time of rapid and much more
widespread reforms, and the differences across countries in terms of the extent of
economic liberalization became less marked, especially during the first half of the
1990s. Both of these studies show that the broadest, most widespread reforms took
the form of trade and financial measures. There was, as will be discussed in greater
detail later on, less uniformity in terms of privatizations and the much more
limited labor-market reforms that were undertaken.
These two phases of the reform process also differed in terms of the types of
changes that were made. This can be shown particularly clearly by an analysis of
trade reforms, which were actively pursued during both phases. In the first, with
the notable exception of Chile, it was mainly a question of streamlining and
reorganizing the unwieldy system of tariffs and quantitative import restrictions
that had been inherited from the era of state-led industrialization. In stark
contrast, the second phase saw a swift and radical reduction in the level and
range of tariffs and the virtual elimination of quantitative import restrictions.
Countries that had large export subsidies also reduced them. During the second
phase, measures to open up trade were also implemented at a rapid pace (in the
space of between one and three years). This meant that, in an ironic twist of
history, the trade liberalization program implemented by Chile in the 1970s,
which was seen as extremely fast-paced at the time, ended up being "gradual"
when viewed within the framework of the patterns that took shape in Latin
America later on.
The differences between these two periods also reflect opposing conceptual
approaches. During the first period, a moderate, gradual approach was taken to
opening up the economy to trade. It was felt that it was better to boost export
growth first in order to prevent the subsequent liberalization of imports from
having a recessionary or otherwise adverse impact on the balance of payments and
to give the production apparatus enough time to adapt to the new policies. During
the second period, on the other hand, the prevailing view was that speed was of the
essence in order to ensure that the process would not be rolled back later on.
A gradual approach was retained only in a few specific sectors, such as some
sensitive agricultural products and, within the framework of South American
integration processes, the automotive industry.
The proposed reforms also varied over time in terms of the degree of macro-
economic discipline that they involved and the extent to which they called for
market forces to be liberalized. On the macroeconomic front, the idea that came
into vogue in the 1970s and especially the 1980s was that it was essential to "get
the prices right." This expression was used to refer, in particular, to placing the
exchange rate at an equilibrium level and allowing interest rates to reflect market
forces. It was also used to refer to the need not to discriminate against agricultural
producers via price regulation by the state, and to the need to set utility rates at
216 Economic Development of Latin America
levels that would cover the costs of the utility companies, which were generally
still state-owned enterprises. Later on, the focus shifted to keeping inflation
low under the stewardship of preferably autonomous monetary authorities.
Under many of the countries' anti-inflation plans, however, the objective of
slowing the rate of price increases was achieved by using the exchange rate to
"anchor" prices. This was done either by reining in the exchange rate or by
freezing it outright, which of course caused the countries' currencies to become
overvalued, in blatant contradiction to the tenet of "getting the prices right" and,
hence, of the idea of correcting any "anti-export bias" or "anti-agriculture bias" of
overvalued exchange rates.
Keeping inflation low also entailed maintaining public finances on a sound
footing, which proved to be more difficult to achieve. In the 1980s, this task was
understood as involving reductions in public-sector spending and, hence, reorder-
ing the corresponding priorities. Fiscal adjustments were made on a massive scale:
national governments cut public spending by an average of slightly over five
points of GDP throughout the decade, which amounted to somewhat more than
one fourth of central government expenditure (see Figure 5.7). Keeping public
finances on solid ground was also seen as involving the improvement of the tax
structure, which, in turn, was understood, for a long time, as equivalent to raising
value-added taxes and lowering direct tax rates. Starting in the late 1990s, this
effort also came to include the establishment of explicit fiscal targets of various
types (e.g., targets for the primary surplus or a balanced budget), together with
caps on increases in public spending, as part of broader packages of fiscal
responsibility rules that also applied to regional or local fiscal authorities in federal
or decentralized systems.
As noted above, from early on the reform agenda included trade liberalization
and, with it, integration into the world economy on the basis of existing compara-
tive advantages, as well as, with very few exceptions (notably Mexico's oil indus-
try), steps to open up the economy to foreign direct investment. The objective of
setting low tariff rates was thus achieved to a much greater extent during this
period than during the classic commodity-export-led growth stage. A wave of
free-trade agreements was also ushered in during this period under the leadership
of Mexico and Chile, which represented a break with more orthodox approaches
(see below). Another important step taken at this time was the signing (by, among
many others, all of the Latin American countries) 11 of the Marrakesh Agreement
in 1993 to create the World Trade Organization (WTO). This agreement not only
expanded existing trade disciplines, as an extension of the GATT, but also created
new ones in the areas of services and intellectual property rights. Trade liberaliza-
tion was also accompanied by the dismantlement of systems for state intervention
in the production sector that had been set up during the preceding stage to
promote the development of manufacturing and agricultural industries. This
approach was embodied in a slogan heard in many quarters: "The best industrial
policy is no industrial policy." In their application of this precept, however,
11 This was not the case with the GATT, which was joined early on only by Brazil, Chile, Cuba, the
Dominican Republic, Nicaragua, Peru, and Uruguay. Argentina joined in 1967, Colombia in 1981, and
Mexico in 1986. The rest of the Latin American countries did so only in the run-up to the creation of
the WTO or when they actually signed the new agreement.
Turning Back to the Market 217
A. Revenue and Expenditure(% of GOP, simple average)
24.0
23.0
22.0
21.0
,.20.0 \1--
'.19.0 1" .
I\
' , '_,18.0 ,.-...,..,.,.. I
\ I
-,17.0 .........
..16.0 \I ,...... I,,"~
,-'.,.- ...."','
.,.,t·
15.0
14.~0~~~~~~~~~~$~~~~~~~$~$~#~~~~~~#~ ~~ (\ #~
- - Revenue (OxLAD) - ·Revenue (ECLAC) - - Total expenditure (OxLAD)
- - Total Expenditure (ECLAC)--- Primary Expenditure (ECLAC)
B. Fiscal Balance (simple average)
3.0, .2.0
\
I\
,'1.0 \
o.o -l--.---,-,,......,.--~~~-.--.~,---,---.--.~~---,-,~~---,-,,......,.---7-~r~~
-1.0
-2.0
-3.0
-4.0
-5.0
~0~~##~~~~~~~~~#~#~(\#
- Deficit (OxLAD) - Deficit (ECLAC)-- Surplus or primary deficit (ECLAC)
Figure 5.7. Central government finances
A. Revenue and expenditure(% of GDP, simple average)
B. Fiscal balance (simple average)
so11rce: OxLAD and ECLAC. The data from OxLAD exclude Bolivia and Cuba; that for Brazil refer to primary
expenditure.
The data from ECLAC exclude Cuba, El Salvador, Guatemala, Paraguay, and Panama.
policymakers left aside one aspect of state intervention about which there was a
stronger consensus: technology policy. This was an area in which very little had
been done during the preceding stage of development. The policy measures
implemented in this field were focused primarily on designing schemes to s.upp~rt
existing demands, and entailed making various funds available that were pnmarily
218 Economic Development of Latin America
taken advantage of by firms that were already innovating, which meant that
sectoral priorities as such did not enter into the picture.
The steps taken to open up the economy to trade were accompanied by the
elimination of most foreign exchange controls and by the liberalization of domes-
tic financial markets. Action in the latter area included the liberalization of
interest rates, the elimination of most of the forms of directed credit that had
been set up during the preceding period, and the reduction and simplification of
bank reserve requirements. The privatization of a wide array of state-owned
enterprises was the third item on this structural reform agenda, along with the
move to open up public utilities to private investment. In this latter case, however,
the process was more gradual and piecemeal, since a number of countries chose
not to privatize state enterprises operating in certain sectors (see below). The more
widespread deregulation of the private sector (including, for example, the elimin-
ation of price controls, the reduction of red tape, and a lowering of entry barriers)
was finally placed on the agenda, although there was recognition of the need to
adopt regulatory schemes for privatized public utilities, to pass broader anti-trust
legislation, and to strengthen financial regulation and oversight (referred to as
prudential regulation and supervision) in order to prevent financial institutions
from becoming overexposed to risk, which would have put deposits of the public
and the sta~ility of national financial systems (systemic stability) in jeopardy.
However, _this new regulatory_agenda was put into practice at a slow and irregular
pace and, ill the case of financial regulation and oversight, only in the aftermath of
severe domestic financial crises.
Social issues did not figure prominently on the initial market reform agenda.
The only mention of spending on education and health made in the original
decalogue prepared by Williamson, for example, was as public spending priorities.
However, th~ widely publicized social-sector reform proposals advocated by the
World Bank ill the 1980s and beyond featured three main ideas: decentralization,
targeting public social spending on the poorest groups in the population, and
opening up opportunities for participation by private-sector agencies and organ-
izations in the delivery of social services.I2 The first of these concepts was part of
an agenda for political reform and should therefore not be regarded as one of the
market reforms. In the realm of social policy efforts, there was in any case an
awareness of the key role of the state and, in fact, a call for public-sector spending
to focus on that front. Reforms of pension systems inters~cted with efforts to
ensure that fiscal accounts were sound. The individual capitalization scheme
introduced by Chile in the early 1980s to take the place of the old pay-as-you-
go system was taken up by other countries in the region and elsewhere (including
a number of central and eastern European countries). Not all pension reform
efforts moved in this direction, however.
This wave of reforms also coincided with the re-emergence of alternative lines
of reasoning that influenced the form taken by those reforms. ECLAC's Changing
Production Patterns with Social Equity (ECLAC 1990) was a landmark analysis
and was to be followed by many other studies in the ensuing years (see Rodriguez
12
For a review of the main ideas that were being formulated concerning social policy, which marked
a sharp contrast with the earlier approach, see Filgueira eta!. (2006).
Turning Back to the Market 219
2006). Outside ECLAC, the resurgence of more heterodox thinking came to be
known as "nee-structuralism" (see, for example, the compilation ofSunkel1991).
The new proposals being put forth revolved around four main ideas: (a) more
active,. countercyclical macroeconomic policies would help to ward off imbalances
during the upward phase of external financing cycles and allow more leeway for
expansionary policies during the downswing; (b) measures for opening up the
economy to the rest of the world should be combined with open regionalism;
(c) active production and technology policies should be designed for the countries'
newly open economies in order tb promote innovation and system-wide competi-
tiveness; and (d) social equity should be placeq at the center of development
(see, in particular, ECLAC 2000; Ffrench-Davis 2005; and Ocampo 2004a).
In time, a number of these objectives came to figure on the agenda of the insti-
tutions promoting the reforms, particularly the World Bank. This was the case
with the role to be played by social policy and technology policy and, following
the international financial crisis of 2007-8, by countercyclical macroeconomic
policy, when the latter principle was embraced by the IMF. Chile was the only
Latin American country that adopted clearly countercyclical fiscal rules prior to
that crisis, in 2001.
The diversity of views and the influence they exerted were reflected in the
macroeconomic management models that were put in place and in the scope and
speed of some of the structural reform programs. On the macroeconomic front,
the chief innovations were the heterodox anti-inflationary adjustment programs
that Argentina, Brazil, and Mexico experimented with in the late 1980s (of which
only Mexico's turned out to be successful in the long run), along with reserve
requirements on capital inflows that were introduced by Chile in 1991 and
Colombia in 1993. In a number of countries, there was resistance to the more
radical forms of economic liberalization and in some there was open (and
successful) political opposition to the privatization of state-owned enterprises
(Costa Rica and Uruguay). In others, the process moved forward but various
enterprises, especially public utilities and oil and mining firms, remained in the
state's hands. In Mexico, nationalization of the oil sector remained protected by
the Constitution. In the financial sector, many development banks and state-run
first-tier banks survived as well. In fact, only three countries took a truly radical
approach to privatization: Argentina, Bolivia, and Peru. It is telling that even the
country that championed these reforms early on-Chile-held on to its state-owned
copper and oil enterprises, as well as its development bank and a government-run
first~tier bank. As a result, despite the privatization of many firms, the share ofstate-
owned enterprises in non-agricultural GDP hardly declined at all in the 1990s; the
public sector's share of the financial market did shrink, but nonetheless remained
substantial in a number of countries (see Table 4.10).
Many of the more radical structural reforms of the 1980s were actually more
of a side effect of short-term policies adopted in order to deal with the crisis
rather than a clearly defined long-term strategy. Thus, the macroeconomic crisis
provided leverage that allowed policymakers to evade political opposition to
structural reforms.
There were also a number of fairly widespread instances in the region in which,
in a sense, the measures being applied ran counter to the more neo-liberal schools
of thought. The most noteworthy cases were public-sector social spending and the
220 Economic Development of Latin America
labor market reforms. In the first case, the trend in social spending was more
favorable than it was for public-sector expenditure as a whole during the lost
decade (although trends differed considerably across countries) and, starting in
the 1990s, a general upswing in social spending pulled total public spending along
in the same direction. In order to hold fiscal deficits down to moderate levels, this
expansion in spending was coupled with rising taxation (see Figure 5.7). In the
second case, the deregulation of the labor market was quite limited when the
reform movement was at its height, and various countries reintroduced stricter
labor regulations in the early years of the twenty-first century (Murillo, Ronconi,
and Schrank 2011). Both of these cases are a telling reflection of the convergence
of economic reforms and the return of democracy to the region.
Another element that was clearly political in nature was the support provided
for regional economic integration, which was in direct contradiction to propon-
ents of orthodox views who were calling for unilateral trade liberalization. The
landmark events in this process were the creation of MERCOSUR in 1991 and the
simultaneous revitalization of the Andean Community and the Central American
Common Market, which had virtually collapsed in the 1980s. Trade liberalization
thus took the path of "open regionalism," to use the term coined by ECLAC
(1994). Mexico and Chile, the two larger economies that were not part of any
formal integration agreement, led the way in signing free-trade agreements with
countries in the region, but they also ushered in a "neo-orthodox" approach to
trade liberalization by signing free-trade treaties with industrialized countries as
well (the traditional orthodox view being that entering into such agreements
would generate trade distortions).
The first such trade agreement was the North American Free Trade Agreement
(NAFTA), which was signed in 1993. It was an innovation inasmuch as, in
addition to deepening provisions on sectors that were already being addressed
by WTO (services and intellectual property), it also covered new areas such as
investment standards and government procurement rules. The failure (which was
never publicly acknowledged) of the negotiations aimed at creating the Free Trade
Area of the Americas (FTAA) that had been launched at the first Summit of the
Americas, held in Miami in 1994, led to a deep division between the countries that
ended up negotiating bilateral (or, in the case of the Central American countries
and the Dominican Republic, plurilateral) free-trade agreements with the United
States and those that refused to do so (the members of MERCOSUR and some
Andean countries). This division also ended up triggering a severe crisis in the
Andean Community in the mid-2000s. Negotiations with the European Union
have moved ahead quite slowly (only Mexico and Chile managed to bring them to
a successful conclusion early on), and, more recently, agreements have been
signed with some Asian countries.
. The diversity in approaches became even greater in the early years of the
twenty-first century, largely because of the triumph ofleftist political movements
that are opposed to market reforms. The "lost half-decade" associated with
the crisis that erupted in East Asia in 1997 was a turning point. From then on,
not only Latin America but the entire world, including international financial
institutions, have exhibited greater pragmatism and have placed new issues on the
agenda, especially those having to do with social equity and institution-building.
The messianic tone in which the reforms were announced (Balassa et al. 1986;
Turning Back to the Market 221
Edwards 1995) and the initial overly optimistic assessments of them-which coin-
cided, ironically enough, with the beginning of the "lost half-decade" (Wodd Bank
1997; IDB 1997)-gave way to a more critical tone and much more nuanced
views.13 Nevertheless, very few of the liberalization measures adopted under the
aegis of the Washington consensus have been rolled back, even in countries that
have returned to more interventionist economic management schemes.
In fact, there is a great deal of divergence among the leftist governments that
came to office during the first decade of the twenty-first century (and the
government ofVenezuela, which came to office in 1998).14 Some combine a fairly
orthodox macroeconomic approach with more active social and, especially, labor
policies, which include large increases in the minimum wage, support for the labor
union movement, and even a return to tripartite wage negotiations (Brazil and
Uruguay). Other left-leaning governments take a more clear~y statist appr?ach
(whether in terms of property ownership or strict regulation and taxatwn),
especially in hydrocarbons sectors (Bolivia, Ecuador, and Venezuela?. Be that ~s
it may, even in the case of the country that has taken the most aggressive stanc~ m
expanding the role of the state (Venezuela), nationalizations were slow in co.mmg
and have been more piecemeal (only beginning in 2009, apart from a senes of
more strategic nationalizations in 2007) than the socialist (o~ even developm~n
talist) models of the past. Bolivia and Venezuela have also mtroduced agranan
reform programs. (Brazil has done so too, but its agrarian reform effort began
before a leftist government took office.) Venezuela and Argentina are the on~y
ones that have diverged to a significant degree from orthodox macroeconomic
policy.
In general, the leftist movements (except, partially, those in Argentina, Ecuador,
and Venezuela) have not rolled back their countries' trade liberalization programs
and in Brazil, this program has been combined with a return to active industrial
policies. The point on which the Left is most clearly in agreement is its emphatic
rejection of free-trade treaties and its support for regional integration. There are
gaps in this consensus, however, as illustrated by the friction in bilateral trade
between Argentina and Brazil and by Venezuela's decision to withdraw from the
Andean Community and to promote a number of specific cooperation initiatives
(many of them bilateral) within the framework of the Bolivarian Alternative for
the Americas (ALBA).
An analysis of the reform process indicates that the effort t.o open u~ the
countries' economies to the rest of the world has perhaps been 1ts most Widely
embraced and enduring feature. It is also the area in which the reform process
most clearly overlaps with changes associated with the second wave of globaliza-
tion. It therefore warrants a closer look.
13 See, in particular, Kuczynski and Williamson (2003) and World Bank (2006), as well as_the r~view
of the debate concerning the Washington consensus of Birdsall, de Ia Torre, and Valencia CaiCedo
(2011). varying analyses of this topic, see, in particular, Levitsky and Roberts (201~) and !~~s1.e
14 For
(2009). As noted by Levitsky and Roberts, in political terms there has also been an ~v1dent d~v!s!On
between countries that accept the ways in which liberal democracies work (Argentma, Brazil, and
Uruguay) and those that seek to recast their political institutions, which are for the most part
governments in which there is a greater concentration of power (Venezuela and, to a lesser extent,
Bolivia and Ecuador).
222 Economic Development of Latin America
INCREASING INTEGRATION INTO
THE WORLD ECONOMY
Trade liberalization has led to significant changes in the way in which the Latin
American economies are positioned in the international economy: they are much
more open now than any time in the past; their goods and services export
structures have been altered, although in different ways in different countries
and subregions; and their manufacturing and agricultural sectors have been
restructured in order to adapt to the liberalization measures that were put in
place and, in many cases, this has led to the disappearance of certain firms and
branches of production. Intraregional markets have increased their share of total
trade, although they are subject to sharp cyclical fluctuations. Transnational
corporations also have a much greater presence than in the past, while the more
successful Latin American firms have expanded into other parts of the region and
some have become world-class players (the so-called "trans-Latins").
At the aggregate level, the most important outcome of the region's trade
liberalization movement has been the increase in its economies' openness to the
outside world. The coefficients used to measure the degree of openness began to
climb in the mid-1960s, and that upward trend started to steepen in the mid-
1970s. The corresponding calculations are shown in Figure 5.8; they exclude
Venezuela, whose oil exports have been strongly influenced ever since the 1970s
by OPEC's mechanisms for controlling the supply of crude oil. The upward trend
in export coefficients has been interrupted only by the deep worldwide recession
of 2008-9, but imports have exhibited a pronounced cyclical trend, with sharp
30.0% ,--~~~~~~~~~-------~--~~~~~~--~~--
25.0% ~-----~~~--~~~~~--~~~~----~----~--~-
\\ /
::::~------------------ y---./~-
;/
100%~~
5.0% ~----~~~-~-~~~~~~~~~~----~~~~~~~-
0.0% ~~~~--~::~::~-:~:o:-o::~::~:~--~-:~:-~:~-=~~~g=-m=-~-~=-~-=~--~=~--~~-~--8:o--~--8:-~~~~~~~~~~~~~~~~~~-
mmmmmmmmmmmmmmmmmmmmmmmmmooooooooooo
~~~~~~~~~~~~~~~~~~~~~~~~~N~NNNNNNNNN
1 - IExports - - · Imports
Figure 5.8. Trade openness (% of GDP at constant 2000 dollars)
Source: ECLAC historical series. Excludes Venezuela.
Turning Back to the Market 223
increases during booms and downswings or stagnation during the three major
crises of the period: the "lost decade" of the 1980s, the "lost half-decade" in the
closing years of the twentieth century and the first years of the twenty-first, and
the recent severe, worldwide recession. In the preceding chapter, Figure 4.2 shows
that the degree of export openness achieved at the turn of the century was
comparable to the levels attained in 1928-9 and has consistently surpassed
those levels during the first decade of the twenty-first century. Thus, the Latin
American economies can rightly be described as being more open to trade now
than at any other time in their history.
This pattern can be seen throughout most of the region (see Table 5.2). The
major exceptions are Venezuela among the larger countries (in the case of its
export but not its import coefficient), and three smaller countries (Honduras,
Panama, and the Dominican Republic) that already had quite open economies by
the mid-1970s. The increase in the degree of openness has, however, differed quite
a bit from country to country, with Chile and Mexico, among the larger countries,
and Costa Rica, Honduras, and Paraguay, among the smaller ones, exhibiting the
most notable increases in the export coefficient.
One fact that has been overlooked in many analyses of Latin American trade
liberalization is that the sharp upturn in the countries' trade openness coefficients
is a result both of swift export growth and slow economic growth after 1980-an
issue that will be explored in greater depth in the following section. As noted in
the preceding chapter, export growth began to speed up in the mid-1960s-and
even earlier in some of the smaller economies-when the classic inward-looking
development model was replaced by a "mixed model" that combined import
substitution with export promotion and regional integration. This acceleration
is depicted in Table 5.3, which shows that, if Venezuela is factored out of the
calculations, the real growth rate for the region's exports of goods and services
reached 8.6 percent in 1974-80. The growth rates for this period are skewed by the
steep upswing in Mexico's oil exports, but even so, the region's simple average
annual export growth rate (without including, here again, Venezuela) was over 6
percent from the 1960s on, declining noticeably only during the lost decade.
In any case, the export growth rates registered for 1990-7 and 2003-8 surpassed
those achieved before the lost decade. The rate did, however, slow during the crisis
that broke out at the turn of the century as a result, as we will see, of the behavior
of intraregional trade. This appears to indicate that the reduction in the "anti-
export bias" of the trade protection system had only a slightly stronger effect on
exports than the greater "neutrality of incentives" that was achieved under the
mixed model, and was clearly weaker in terms of its effectiveness in spurring
dynamic economic growth. This backs up the conclusion, as stated earlier, that the
sharp increase in the countries' coefficients of openness up to the early twenty-
first century was just as much-if not more-a reflection of slow economic growth
as it was of stronger export growth.
The expansion of exports was accompanied by shifts in the export structure as
part of a process that was taking place throughout the developing world. 15 The
reasons behind those changes at the world level included: a steady build-up of
15 See Lall (2001: ch. 4), Akyiiz (2003: ch. I), and Ocampo and Vos (2008: ch. III).
00 \C?ll1~tt'!~~t'--:
0 ON0001""""'100
I 1""""'1....-t('t')Nri')NN
"00 '
"'
"0I'
00
"~'
e0 "".!'.
.§ "~'
0
".!.'
00
~
0
00
I
lt)
"~
00 NI..OLOOO'\t'--0
0 ~~Lrir--:oOa\M
....... .....c("().....cN.....cN
I
"00 '
"'
"0I'
00
~"'
e0 "".!'.
"~'
~
0 0...-1('()1.1)~00 0\'<:f'IL0~0~\00\0\0l/)
".!'. l.d..Orria\000 o\a\N~c-r>.....:Nc.Ot-.:\C5N
00 N 1""""'1.....cN l"""''('t").......C.....c("()NI..().....c.....c.....c
~
0 - --lr}......tLOOO.....cNN -\0['-...~("C')-l.00\\0\0N
00 ,....:I.C)v)~rt"it....:r...:o\,....;Lr),....:
~~r--:oO\ciOo\ .....c.....c("().....c.....cN("()('(').....c.....t.....t
I
lt)
~"
Turning Back to the Market 225
Table 5.3. Real growth of exports of goods and services (average annual growth at constant
2000 dollars)
1960- 1967- 1974- 1980- 1990- 1997- 2003- 2008-
1967 1974 1980 1990 1997 2003 2008 2010
Weighted average 4.6% 6.0% 8.6% 5.5% 9.5% 6.2% 6.5% 0.8%
Excluding Venezuela, 4.1% 3.9% 5.5% 5.2% 9.0% 5.4% 6.2% 0.3%
Latin America
· (18 countries) 3.7% 3.2% 5.6% 5.7% 9.0% 5.5% 5.9% 0.3%
Large and medium-
sized countries (7) 6.7% 7.7% 5.2% 2.7% 9.4% 4.7% 7.8% -0.2%
Small countries (ll)
Simple average 6.1% 6.1% 6.2% 3.9% 8.7% 5.6% 7.9% 0.3%
Excluding Venezuela, 5.9% 5.7% 5.6% 3.9% 8.5% 5.1% 7.5% -0.4%
Latin America
(18 countries) 3.8% 3.7% 7.0% 4.6% 8.1% 4.8% 6.4% -1.4%
Large and medium-
sized countries (7) 7.3% 7.0% 4.6% 3.4% 8.8% 5.3% 8.3% 0.3%
Small countries (11)
Source: Authors' estimations based on ECLAC historical series
production and technological capacities, especially in East Asia; the breakdown in
production (or value) chains that made it possible to divide up the different
steps in the production of a given product (including its final assembly) and
spread it out over different locations; the strong incentive that, under these
conditions, sharp wage differentials create for boosting profits by moving the
more intermediate- and low-skilled labor-intensive processes to developing coun-
tries; and the growth of trade in services, which has, in many instances, been
spurred by the revolution in information and communications technologies
(ICTs). Transnational corporations have played one of the most important roles
in this respect, both because of their control over value chains and because of their
active participation in the internationalization of services, which has entailed the
maintenance of a trade presence in other countries.
Generally speaking, commodities and natural-resource-based manufactures
have tended to decline as a share of developing-country exports, while the share
of low-, mid- and especially high-technology manufactures has been on the rise.
In the case of assembly operations, it should be taken into account that production
processes are technologically simple, even if the goods involved are classified as
being mid- or high-technology products (e.g., automobile assembly and computer
assembly, respectively).
As a result of the fragmentation of production processes, countries also import
many of the parts and components. When they are assembled in free-trade zones
using diverse tax incentives (including income tax incentives), the associated
activities can accurately be described as "enclave" industries. These industries
may be highly sophisticated in terms of the goods that they produce-or, more
often, merely assemble-but they remain enclaves.
Another characteristic of international trade in recent decades has been the
increase in trade flows between developing countries. Most of these flows are
intraregional, but some are interregional. Examples of the latter include trade
226 Economic Development of Latin America
Table 5.4. Latin America: Compositions of exports of goods, 1990-2008
1990 1997 2003 2008
Commodities 51.1% 31.8% 30.1% 38.0%
Resource-based manufactures 19.7% 17.4% 15.7% 19.1%
9.2% 8.5% 7.9%
Agricultural 10.5% 8.7% 7.2% 11.2%
Other 70.8% 8.7% 45.9% 57.1%
Subtotal resource-based goods 9.6% 49.1% 12.0%
Low-tech manufactures 5.2% 12.5% 5.6% 7.6%
Textiles, apparel, and footwear 4.4% 6.4% 6.3% 3.0%
Other products 15.8% 6.1% 25.4% 4.6%
Medium-tech manufactures 4.2% 24.6% 9.5% 20.9%
Automotive industry 6.1% 9.2% 5.1% 7.6%
Processing industries 5.6% 5.3% 10.7% 5.7%
Engineering industries 2.6% 10.1% 14.8% 7.7%
High-tech manufactures 10.8% 12.5% 11.4%
Electronic and electrical 1.5% 9.4% 2.3% 9.4%
Other 1.0% 1.5% 40.2% 2.0%
Subtotal medium and high-tech 18.4% 35.5% 2.0% 32.3%
Other 1.2% 2.9% 3.0%
Source: Authors' estimates based on UN-Comtrade
between Latin America and China and, to a lesser extent, other East Asian
countries. As we will see, these two types of "South-South" trade have had
radically different impacts on Latin America's export structure. Since 2004,
another factor in this respect has been the surge in primary commodity prices
(driven mainly by the dynamism of China's economy), which has had a wide-
spread impact in those parts of the developing world whose production structure
is based on natural resources.
Along with the rest of the developing world, Latin America has been witnessing
a transformation of its export structures. As shown in Table 5.4, between 1990 and
1997 the share of commodities shrank and the level of exports of intermediate-
and high-technology manufactures grew quite rapidly and then continued to
climb, although more slowly, during the crisis that occurred at the turn of the
century. Between 2003 and 2008, just the opposite occurred, as the share of
commodities and natural-resource-based manufactures expanded. This latter
trend is no doubt driven, at least in part, by the high level of commodity prices.
This "re-commoditization" of the region's export structure stands in contrast to
the continuing diversification of the Asian countries' exports.
This process has followed different paths in different parts of the region and has
tended to give rise to two basic patterns of specialization that broadly follow a
regional "North-South" division (ECLAC 2001a; Ocampo and Martin 2004;
Ocampo 2004a: ch. 1). The "northern" pattern is one of rapid diversification
toward exports of manufactures having a large component of imported inputs
(and including, in its most extreme form, maquila industries) which are primarily
destined for the US market. In the Central American countries, this pattern is
found in conjunction with a growing tourism industry and, in a number of
these nations, with significant commodity and natural-resource-based manufac-
turing sectors. The "southern" pattern, which has, in contrast, changed less
radically in recent decades, is made up of a combination of extra-regional exports
Turning Back to the Market 227
of commodities and natural-resource-based manufactures (many of which are
also capital-intensive) and a much more diversified range of products (including
many with a greater technological content) that are traded within the region.
Brazil is positioned somewhere between the two groups, since it had a much more
diversified export structure (including some technology-intensive manufactures
and services) than the other South American countries before liberalization
occurred,16 but that structure has changed relatively little since then. In addition
to these two basic patterns of specialization, a third pattern can be discerned in
Panama and some Caribbean countries (the Dominican Republic and Cuba,
which are not included in the table), which are primarily service exporters.
Table 5.5 illustrates the patterns (including the service exports pattern) found in
individual countries. An analysis of this table brings out some significant aspects
of these broadly defined patterns. Focusing first on exports of goods, it can be seen
that the most common shift has been toward the diversification of natural-
resource-intensive manufactures. This pattern has been quite marked in about
half of the countries of the region, with Chile and Peru, in South America, and El
Salvador and Honduras, in Central America, being the most notable examples.
The next most common pattern has been a shift toward mid-technology manu-
factures, with the specific types of products varying from one country to another.
Mexico, Colombia, Argentina, and Costa Rica are the most conspicuous examples
in this case. In contrast, diversification toward low-technology goods has generally
not taken place; instead, the shares of exports of textiles and apparel have been
shrinking in a large group of countries. The main exceptions are El Salvador and
Guatemala, where exports of those products have grown and represent a major
export category. Only Mexico and Costa Rica have made any significant headway
in exporting high-technology products, although, in both cases, much of this
corresponds to assembly activities.
Table 5.5 also illustrates the heterogeneity exhibited by both South America and
Central America. In South America, after Brazil, the countries with the most
diversified export structures are Argentina, Colombia, and Uruguay, while the
other countries in that subregion are much closer to displaying a pure version
of the "southern" pattern. In four of them, commodity exports continue to
account for more than half of all external sales (Bolivia, Ecuador, Paraguay, and
Venezuela), while in Chile and Peru, as noted earlier, natural-resource-based
manufactures are becoming increasingly important. In Central America, the
situation is even more varied: Costa Rica and El Salvador exhibit the greatest
degree of diversification, while Nicaragua is at the other end of the spectrum. It is
interesting to note that the countries that made the most progress during the
period of state-led industrialization now have, for the most part, more diversified
export structures. This is the case of Brazil and Mexico, followed by Argentina and
Colombia, among the larger countries, and of El Salvador and Guatemala, all?-ong
the smaller nations. Costa Rica has clearly been the most obvious winner in terms
of diversification. But diversification toward activities not based on natural
16 In fact, in 1990 Brazil's and Mexico's export structures were not very different in terms of this
classification, other than that natural-resource-based manufactures represented a larger share in the
former while tourism services figured more prominently in the latter.
"#'#.'#. ~ #,(.f. (fl. ?f.?F.?f!.Cf.
""0
""N 0 o.-<cNr-U--'>: o0\..~..\.0:o1.</")i
0 Ot-..'0 N
~N\0 "' <f. <f. <f.??.
t"r?t').N'..#....';#"o"->\ ..,. "''#.??.'#. ~\ONN
"" ,....;..........;N NN,....;I.ci
??.??.??. '#.??.'#.'#.
"'0 0 "
t.r)c)C1") NO\O......c
#.??.??.
"' OONI..Cir'
o*R"'"'??.??.??. d"'d"c"O"'
" .-< N
.....
??.??.??.??.
Cr-O-:O...[..':N.,....~.;
.-< N
O"'#".C'd0§(?d"f'. <ofr.--.<..f.....<.cf-..?::?tt.
"" o,....;_;o\
<f.??.??. ' N .-<
t-..0'<1' <f. <f.??.??.
ddo\ 1.1")01'"""1\0
" t<)NLI"id
.-<N
??.<f. <f.??.
OOt--.0\D
,....;t'f").....i\0
??.??.??. ??.'#.??. <f.??.??.??.
c.-<oNoN .-<Nt'"l 0~('1"')1:'--,
"" NrriO ffi~No\
??.??. ??.??.??. ??.??.??.??. "' ?1?.'#.??.??.
":"1"1
t..nOON'O:fl ??.#.??. 000\I.O......C
N0 0 .....:r---.:~M
o",".."..;"o' Lri~OO~
"" ..... N 00
?f!.??.'#."#-
00 "~'~"' <f. <f.??. .s
0 ('t)- tv){'..
0 U'> U'> N ]
N ??.??. ociN~~
0 .-<t-.. c?rrif;j
§ """C)0 r...:~ '#.??.??.
N
0"' N
c),....;rr) ??.??.<f.<f.
"' 0\...-10000
\C)t.r)u)~
?f.??.?ft-
'.#.,..?..,f..?.?.... ?r-?-.-<.tf...n<r-f-.-?.0?\. "00" d0 d...\... 0""' '#.Cf.Cf.??.
"o \ 0 0 N~cio\ N 00 0\t--..Nco
.-<.-<f'l
..... OMN..O
??.<f.??.??.
??.<f.??. <f. <f. <f.??.
1..000\l.()
N ""> U'> N..O~rri Nt'-.0000
Nu)\.0~
Nl.cici .-< N
??.??.<f.<f.
"
00..-t\Dt.r)
??.'#.??.
r..:\ON~
U'>'O.-<
~,....;.ql
N "'
??.<f. <f.
t"1"")"c)'\"C)'
"
Turning Back to the Market 229
resources has been the exception rather than the rule. As of2008, only a handful of
countries (Mexico, Costa Rica, El Salvador, and the Dominican Republic) had an
export structure in which natural resources and related manufactures accounted
for less than half of all goods exports.
Service exports have been less robust and, in fact, the region's share of world
trade in services has declined over the past two decades (see ECLAC 2007b:
ch. III). National patterns vary widely, however. Three economies display the
third, service-based pattern of specialization: Panama, which exports transport
and financial services, and Cuba and the Dominican Republic, which export
tourism services. Tourism exports have been expanding in almost all of the
Central American economies as well. Brazil and, to a lesser extent, Argentina
and Uruguay have seen strong growth in more technology-intensive service
exports (included in "other services"). Another two South American economies
are also important service exporters: Paraguay (electricity) and Chile (transport
services, in large part as a by-product of its exports of high-value-added agricul-
tural goods). Mexico and the Andean countries have done the least in terms of
positioning themselves within the growing world trade in services, although
Mexico's tourism industry has been strong since the beginning of the period
under review here. One interesting area of activity which has not been touched
upon here, and in which some of the countries of the region have a notable
presence, is the culture industry (music, film, television).
Intraregional trade has made a significant contribution to the growth of exports
of manufactures, but it has also been a factor of great instability in South America.
While intraregional trade came to account for a growing share of the region's
exports of goods between 1990 and 1997, in the cases of MERCOSUR and the
Andean Community that share shrank during the crisis of the turn of the century
and during the severe worldwide recession of 2008-9 and, since the end of the
1990s, has remained below its 1997 level in both of those groups (see Table 5.6).
On the other hand, intraregional trade has been growing steadily within the
Central American Common Market, and the slump in such trade during
the 2008-9 recession was also smaller than the decline experienced in the rest of
Latin America.
Intraregional trade has included a larger share of mid- and high-technology
products than other trade flows even prior to the 1990s, and this has been especially
true for the Andean Community and the Central American Common Market. This
underscores the advantages offered by intraregional trade in terms of industrializa-
tion, particularly for the smaller economies. This pattern became stronger in the
Andean Community and in MERCOSUR during the boom of the 1990s, but this
was not the case in the Central American Common Market, as Costa Rica's exports
of high-technology products to outside markets increased. During the 1990s boom,
all integration processes also offered opportunities for exports of natural-resource-
intensive and low-technology manufactures (with the exception of Central America,
in the latter case, since it continued to export manufactures with a greater techno-
logical content). The relative decline in intraregional trade during the crisis that
erupted at the turn of the century therefore dealt a hard blow to the exports of
manufactures of the two South American integration groups.
One of the most important trade flows between the region and other developing
markets is, as noted earlier, Latin America's growing trade with China. This
00 0"\~f'-,.I.OlON~('t')
0
0 rt")oOff'ici..fo\oOa\
'i:l N
~ ~"d""\OlOI.O ........ I.O('t')
::8 CQ['..I..f)COOOlOO
c:: .....;oQrt)~u)Nrr)r'
0 "0 '
......C~\OI.Olf)("'("()('t')
8 0
8N .....ct-..NfCO\OI'"""'lO
u0 ~OONt...:t.ONoO
a·c::
"a~'- I.Ot.n"=tt\Oo::tl('t')N
OJ ~lf"!~'-qll)lf1'""'!N
~ 1.0~~~~~::!:~
]
c::
uOJ
a0 -
~
00 Q'\ll)'o::f'l('f')0\00\0.....C
0 N~et.S~.....tu)tri..--'
0 .....CI.O('I')t'--\0\0 N.
N
0 f'()['-..00\\0.....CI.OO'\
ffi~o\No\o\t--:~
l "'0 ......c-qt("()~-.:tl\0 N
~N
u0
"~:'; "...'.OJ
aa-- t.n\O"'d''0\0\t--..COt--.
,.....;<'iN~r...:~Off'i
.....c"d"'("()lO\OOO.....CN
0a- ~f.::~~~ci~~
~
.....cN.....c('l')-.:t'l"-.('().....c
t":' 00 OOOOCO"':ttlOOO('t')OO
0 r-..:o\u)M,....t~u)o\
0 .-l('f')N~lO..::r N
N
0
~
<..d... 0 oq<""}C"i~~O':O':C'i
0
"'.0....
~ "'0'+-< oo.....c,....,OOlO\Ol0\0
~""""~f'iNN~N N
0
</)
j~
"' ~ "' ~~gi~~*o\~a- 0')0["-...1..1)\000~0
oj ~
~
t!
0
~ a0 -
~ -.:t'C"""""Nf'...~OONt-..
ff'ici0vir'trio0r'
,.....N......c,.....NN .....c
<cd::
0
"6h
~ .;:)
-"' g dI
tc:l: ""d
;·:~aaj~1jblJg~.)Dv~~v.;.~aci§i:~J·J~i,l§b.~Bo uu.<....
-o :;Jl.3~t:r:!:J]
U0 ::8 O-5t-<0 "-1
tri
H...
Q)
"c
~
V)
E-<
Turning Back to the Market 231
country also competes with Latin American producers in other markets, notably
in the US (Gallagher and Porzecanski 2010). This trade activity has had widely
varying impacts on the region. Its most positive effect has been that it has
provided a market for South America's natural-resource-intensive exports,
although the range of products is quite limited (oil, soybeans, copper and copper
products, and iron ore and scrap iron). On the other hand, the robust increase in
China's exports, mainly of manufactures, to Latin America has saddled the region
with a hefty trade deficit with the Asian giant. This is especially the case for
countries that are not major commodity exporters, such as Mexico. China has also
been a growing source of competition for the region (and, here again, especially
for Mexico) on markets for its manufactures. The benefits for the Latin American
economies have therefore been ambivalent, especially since their trade with China
has also been a contributing factor in the re-commoditization of Latin America's
export and production structure.
The most significant indirect effect has been that China's headlong growth has
driven up world commodity prices, especially since 2004, in the midst of a world-
wide economic boom. The turning point for that trend came in mid-2008 and the
downswing deepened with the severe recession that followed. China's rapid recov-
ery soon caused those prices to rebound. In the case of agricultural products, the
boom that began in 2004was actually more of a recovery (and only a partial one in
the case of tropical agriculture) from the downward spiral in the real prices of such
goods during the two preceding decades. Thus, this was a price boom for energy and
mining products more than for agricultural goods. The big winners were therefore
the countries where the former product categories account for a large share of
output: Venezuela and Chile, followed by Peru, Bolivia, Ecuador, and Colombia
(Ocampo 2007). The growing biofuels market has, however, established a direct link
between the markets for these two groups of products.
Trade specialization and the nature of foreign direct investment (FDI) flows
have been closely interlinked. The "northern" pattern of specialization has
attracted transnational corporations that are active players in internationally
integrated production networks, whereas in South America, investment has
been concentrated in services and natural resources. As part of this process, the
sources of investment shifted as Spanish transnationals moved into the region. In
addition, as mentioned earlier, some large Latin American firms joined the ranks
of world-class transnational corporations. The biggest of these "trans-Latins"
come from Brazil and Mexico, and some of them are active players in global
markets, but there is also a group of smaller trans-Latins from various countries
that play an important role in intraregional markets.
FDI rose sharply in the 1990s and peaked, in terms of the net transfer of
resources, between the mid-1990s and the early 2000s (see Figure 5.2). A large
part of the investments made during this boom involved the purchase of existing
firms (both state-owned corporations, which thus were privatized, and, increas-
ingly, private firms) as part of the global wave of mergers and acquisitions. This
means that FDI made less of a contribution to the accumulation of production
assets than the corresponding financial flows would appear to indicate. Although
inflows have remained high (2.8 percent of Latin America's GDP during the
2004-8 boom), mounting outflows in the form of profit remittances and the
rising level of foreign investment abroad by Latin American firms (and
232 Economic Development of Latin America
entrepreneurs) (1.8 percent and 0.9 percent of GDP during the same period) cut
deeply into the net transfer of resources generated by foreign investment during
the first decade of the twenty-first century.
The emigration oflabor to industrialized countries, especially the United States,
is another prominent feature of the region's new patterns of integration into the
world economy. The flow of Latin American workers to the US, which had
swollen toward the end of the period of state-led industrialization, was turned
into a flood by both push factors (the debt crisis of the 1980s and civil war in
Central America, as well as the "lost half-decade") and pull factors. As a result,
according to US census figures, the number of Latin American and Caribbean
immigrants living in the United States jumped from 3.8 million in 1980 to 7.4
million in 1990 and to 14.4 million in 2000, reaching 18.6 million as of2008; this
figure should be raised by another 40 percent or more to account for undocu-
mented migrantsY People have also migrated to other countries, especially
nations in Western Europe (with Spain, a country that had historically been a
source of emigrants to Latin America, being the most popular country of destin-
ation), Canada, and Japan. The number of Latin American and Caribbean emi-
grants to Spain soared from 0.4 million to 2.4 million between 2000 and 2009
(from 0.2 million to 1.8 million if we exclude those who had Spanish nationality),
and another 2 million people emigrated to other countries outside the region.
Moderate levels of intraregional migration were also seen in the 1990s. Costa Rica
and, to a lesser extent, Chile have become major destinations and Argentina has
continued to be one. On the other hand, the migration of Colombians to Vene-
zuela, which had been the largest migration flow in the Andean region, came
virtually to a halt with the debt crisis and has not resumed since then.
One major result of these events has been the rapidly increasing importance of
migrants' remittances as a source of foreign exchange for Latin America. Remit-
tances climbed from around 0.3 percent of Latin America's GDP in the early
1980s to nearly 2 percent just prior to the recession of 2008-9. The share of GDP
that they represent is much greater in small economies, especially those of Central
America, the Dominican Republic and Ecuador. The 2008-9 recession weakened,
at least temporarily, the pull factors at work and generated a break in this trend,
which was reflected in a decrease in migration flows and a 15 percent drop in
remittances in 2009, and the decline has not yet been entirely reversed.
MACROECONOMIC PERFORMANCE
Albeit with some setbacks in individual countries in later years, the most import-
ant economic achievement of the 1990s was the increase in public confidence in
17 See ECLAC (2006b) and Canales (2011). The data refer exclusively to the countries covered in
this book (and therefore exclude Haiti). The estimates for migration to Spain and other destination
countries are based on data provided by the Population Division of ECLAC. For a detailed analysis of
migration flows from the point ofview of both countries of origin and receiving countries, see Martinez
Pizarro (2011). According to the Pew Hispanic Center, 28 percent of all migrants to the United States
were undocumented in 2010 (see <http://www.pewhispanic.org/>).
Turning Back to the Market 233
the macroeconomic authorities that followed the slowdown in inflation and the
improvement in fiscal conditions. In view of the long history of inflation of some
of the South American countries, the widespread experience with runaway infla-
tion during the lost decade and the sharp fiscal imbalances that troubled the Latin
American economies in the late 1970s and early 1980s, the success in stabilizing
prices and turning the fiscal situation around are clearly noteworthy. At the level
of macroeconomic policy, the acceptance of both of these objectives-low infla-
tion and a sound fiscal position-is quite widespread, as these aims have now been
adopted even by political movements that, in the past, downplayed their import-
ance. The appreciation of the need to keep inflation under control has also been
reflected in the greater power and autonomy accorded to central banks in many
countries.
The headway made in combating inflation has been more uniform and longer-
lasting than progress made on the fiscal front. Even in the early 1990s, single-digit
inflation was still the exception, and the region was yet to experience the final bout
of hyperinflation that was part of the cycle that had begun in the mid-1980s (in
1993-4 in Brazil). Nonetheless, inflation rates fell steadily from the early 1990s on,
and by 1997-and, even more clearly, from 2001 on-single-digit inflation had
become the rule in the region. The major exceptions have been Venezuela and
Argentina and, for a brief period, Brazil and Uruguay (during the sharp adjust-
ments made during the "lost half-decade"). In 2008, the wave of inflation in world
food prices had an adverse effect on many Latin American countries. In any event,
at between 6 percent and 10 percent, average inflation rates during the first decade
of the twenty-first century have been higher than the levels seen in industrialized
countries.
The countries' central government deficits were reduced considerably in the
second half of the 1980s and have fluctuated between 1 percent and 3 percent of
GDP since then. Those variations have been cyclical, with the deficits being lower
during booms (between 1 percent and 2 percent of GDP during the 1990s and
around 1 percent in 2004-8) and higher during crises (around 3 percent at the
start of the crisis of the late 1990s and in 2009). The inroads made in this respect
have, however, been less long-lasting and less widespread than the progress made
in curbing inflation, and nearly all the countries have seen their deficits rise to over
3 percent at one point or another, particularly during crises. In addition, Bolivia's
deficit remained above the 3 percent mark between 1998 and 2005, and Colombia
was in the same situation in 1996-2006. Some countries-notably Argentina and
Ecuador-were overtaken by debt crises in the early years of the twenty-first
century, and public debt levels remained above 60 percent of GDP until the
boom of 2004-8, when they began to fall in most of the countries. Since 2007
they have remained at around 30 percent of GDP or less.
It is important to note, however, that these outcomes have been accompanied
by increases in public-sector spending and, as we will see in the next section, in
social spending in particular. Primary spending (i.e., excluding interest payments
on public debt) by the central government rose from an average of around 13
percent of GDP in the early 1990s to 19 percent in 2008 and to slightly over 20
percent in 2010. From a longer-term perspective, this increase can be seen as a
reversal of the sharp reduction in spending that occurred during the lost decade,
since total public spending in the region, measured as a percentage of economic
1
234 Economic Development of Latin America
10.------------------------
6
4
-2~--------------------------
1 - Weighted average
Figure 5.9. Latin American GDP growth rate, 1951-2010
Source: ECLAC. Excludes Cuba.
activity, is not very different now from what it was before the debt crisis (see
Figure 5.7). In order to fund this increase, which has, in addition, been fairly
widespread, governments have had to find ways to boost their revenues as well.
However, a comparison with OECD figures shows that the region continues to
derive a much larger share of its public revenues from indirect taxes (especially
value-added taxes) and from earnings derived from the exploitation of natural
resources than the industrialized world does, while direct taxation (personal
income and property taxes) and social security contributions still provide a
considerably smaller portion of government revenues (Santiso and Zoido 2011).
These advances stand in contrast to trends in economic growth, which have
departed from the patterns that characterized the period of state-led industrializa-
tion in two important ways: a much sharper business cycle and slower long-term
growth (see Figure 5.9 and Table 1.5). Greater price and fiscal stability has there-
fore been coupled with greater instability in production. This is a reflection of the
Latin American economies' heightened vulnerability relative to the pattern that
was typically seen in the preceding period. The second of these two features
suggests that, with the exception of some countries and of the broader regional
growth surge experienced in the 2000s, the reformers' expectation that greater
openness to the rest of the world and improvements in macroeconomic manage-
ment would lead to more robust economic growth has not been borne out.
Since 1990, two entire business cycles have taken place and, at the time of
writing, the region was in the midst of a third. The upswing of the first of these
cycles was triggered by the renewal of access to international capital markets in the
early 1990s. Net resource transfers via the capital account, which had been
1
Turning Back to the Market 235
negative since the eruption of the debt crisis, became positive once again (see
Figure 5.2). Apart from the temporary stutter caused by the crisis that broke out in
Mexico in late 1994, an ample supply of external credit continued to bolster
economic growth until the outbreak of the crisis in the emerging economies
(which began in East Asia in 1997 and then spread to Russia and to the bulk of
the developing world in 1998) cut short the flow of external financing, apart from
flows of foreign direct investment. The impact of all this on the region was seen in
a sharp slowdown or outright recession in a wide range of economies, especially
those of South America, and another half-decade of lost ground in terms of
economic development.
During the second cycle, the Latin American economies experienced the
2004-8 boom, the strongest since 1967-74. This is even more striking when
viewed in terms of the unweighted average of national growth rates, which
indicates that this growth surge was less dependent on the region's two largest
economies. This boom was based not only on the upturn in external financing, but
also on the unusual combination with windfall profits from commodity exports
and high levels of remittances from migrant workers. This conjunction of positive
factors started to fade as the growth of remittances began to slow down in 2007
and soaring commodity prices began to descend in mid-2008, ultimately giving
way to a strong negative external shock that was associated with the global
financial crisis that erupted in September 2008 and the worldwide Great Recession
that followed in its wake. Even earlier in that year, some of the economies in the
region had already experienced a major slowdown and, by the final quarter, all of
them had seen their growth rates plummet or had entered into an outright
recession. The overall effect was a 2.1 percent contraction in Latin America's
economy in 2009, which was the steepest drop to be experienced since the debt
crisis. This constellation of adverse factors began to ease in mid-2009, and growth
rebounded to a 6 percent rate in 2010. That year can be regarded as marking the
start of a new upward phase, although the region's growth rates fell again in 2011
in the midst of the uncertainty that continued to exist in the world economy in the
aftermath of the Great Recession.
This greater degree of openness to the outside world has therefore made the
region's economies highly vulnerable to external shocks, whether positive or
negative. Macroeconomic policy has often heightened these sharp cyclical swings.
Fiscal, monetary and credit policies have tended to be procyclical in most of the
region's economies. Positive external shocks have tended to be accompanied by
increases in public spending, a more ample supply of credit (along with lower
interest rates), and currency appreciation. On the other hand, negative external
shocks have tended to trigger a decrease in public spending, a tightening of credit
and currency depreciation, along with, until the crisis of the turn of the century,
climbing interest rates. As we will see, this procyclical behavior not only tends to
transmit external cycles to the domestic economy, but also makes countries more
likely to experience national financial crises in the aftermath of booms in external
and domestic financing.
In recent years, some progress has been made in introducing countercyclical
policies, but this transition has been only partial in scope. During the 2004-8
boom, in particul}lr, external borrowing increased less and a greater amount of
international reserves was set aside. In addition, in a few cases (the most
236 Economic Development of Latin America
remarkable one being Chile), fiscal policy was countercyclical, although macro-
economic policy generally remained procyclical (Ocampo 2007 and 2011; IDB
2008b). During the Great Recession of 2008-2009, macroeconomic policy took a
more clearly countercyclical turn, particularly with regard to monetary and credit
matters and, in some countries, fiscal issues as well.
Another effect of cyclical swings has been the instability of the countries' real
exchange rates. This has been a much more serious problem in the South American
economies, not only because of the volatility of capital flows and these economies'
heavy reliance on natural-resource-intensive exports, but also because of their
preference for more flexible exchange rates. As a result, sectors that compete with
imports felt the dual impact of greater external openness and currency appreciations
in 1990-7, which made it more difficult for them to make the transition to the new
development model. In addition, because of the greater instability of real exchange
rates in these economies, it has been harder for export sectors that do not enjoy the
static comparative advantages associated with natural resources to gain ground.
Lower inflation and renewed economic growth in the 1990s paved the way for
progress in financial deepening, which can be measured by looking at the amount
of credit that the financial system provides to the private sector and, more
recently, at stock-market capitalization (see Table 5.7). As may be seen, the
countries' levels of financial development in 1980 were between the averages for
lower-middle-income and upper-middle-income economies (but generally closer
to the latter), which are the categories to which almost all of the Latin American
countries belong. After losing a great deal of ground during the lost decade, they
returned to an upward path and stayed within the above-mentioned range,
although, oddly enough, they once again were lagging behind the upper-middle-
income countries in the first decade of the twenty-first century.
As in the past, the level of financial development differs across countries,
although in different ways than before. Argentina is no longer in a lead position,
nor is Venezuela, which was at the forefront in this respect in 1980. The undis-
puted leaders now are Chile and, to a lesser extent, Brazil. Panama, which is,
strictly speaking, an international financial center, might be added to the list, as
could Honduras in terms of lending activity and Peru in terms of stock-market
capitalization. Financial liberalization has doubtlessly contributed to this result,
but it should also be noted that, in one of the countries with the most highly
developed financial systems (Brazil), government financial institutions continue
to maintain a strong presence, as well as in many other countries. During the
2008-9 recession, the use of these public-sector financial institutions as a tool for
reactivating the economy was an important component of countercyclical policies
in a number of countries in the region and in the developing world in general.
As mentioned earlier, the financial liberalization process unfolded in the midst
of a glaring lack of prudential regulation. Against the backdrop of abrupt external
financing cycles and their transmission to the countries' economies, this trans-
lated into extremely frequent systemic financial crises, as had also been the case
during the debt crisis of the 1980s. In fact, two thirds of the countries of the region
(twelve out of eighteen, if we exclude Cuba) underwent national financial crises in
the 1990s or early 2000s (Laeven and Valencia 2008). These crises absorbed huge
amounts of fiscal and quasi-fiscal resources and impaired the operation of the
financial system. Thanks, however, to the reinforcement of prudential regulation
~('f')o:::f'llOI""""'I""""'("l('f"')
O~cv)N'l)oddt--..:
('lj....C;j~Ln l""""'("f")
\ON\00\NOOO\....C QOll)('f')- (',.
vi~~Ou)~'£)N
r'~rf'i~ ~
I..().....C('f')0\1""""11""""1 - -"' "'
OOOON\ONM\0('()
r--.:Nr'Mr--:'CiOO~
- NCO~""""~
00
0
0
"'
('<) 0\~N('f')O\~OO....clO('f')('f')\OOQ('f')NOO~
"'0 ~~oO~~oOoONNv)vi~~~o00oO
0 f"""''"'1!N('..NN~""""~"=ttN('f')l""""'t--......cNNLn
~~~~~~q~~~~~q~~~~
ONO....c~Nr--..r--..r--.~0('..0\....COLnO
NlO~\O('f"')....CN('f")....c('f')N\ONNNN....c
0N~\ 0~\\0C0)\ ~~N(~'f'~)NO~\~(~'..~('~f')~~('~f')~Lr)q('f~')
....ci""""'N....CN-('()1""""1 NNN
....CI""""'N~
~l / ) ~~~~~~qq~~q~~~~~~
00 N~N('f')N('..\O~OOI""""'I""""'lO("f")N~~O\
~""""~ NlO('f')I""""'N('f")l""""'('f')I""""'~I""""'-N('f")~
~~~~~~~~~~~~~~~~~
OO~lOOl0\0\0~~001..()('..('..0\('..000
I""""'I""""'N('f")NN~""""~N~""""~N~""""~~- NNlO
t--...t--.0\1""""10\000NOON~0'\('()0'\
~'C>Lrii.O~.....;oo~o\r---:OOMO~
NN....cN....cNNlO~""""~~""""~N~""""~('f')
00\00\0\0rtl"o::t'!t'--.O(',.I""""'OlO
r--.:.....;ff'it-.:o\NNOr--:o\u).....;r'\0
NN~""""~~""""~~""""~N("f")N...-ll""""'l""""' N
00\....C~......c('I)O'\ONONN\000
~r'ct:i..Oo\N...-4u-i~ct:ir'r'~~
......cN......c......c......c......cN......c ......c ......eN
238 Economic Development of Latin America
following this series of crises and to the headway made in introducing counter-
cyclical policies, the 2008-9 recession was the first in recent decades that was not
accompanied by any national financial crisis. This stands in sharp contrast to what
occurred in industrialized countries during the recent crisis and in Latin America
itself in the past after periods of booming capital inflows (see Chapter 1).
The fact that economic growth has been slower since 1990 than it was during
the period of state-led industrialization is also illustrated in Figure 5.9. The
average pace of growth in Latin America between 1990 and 2008, which can be
regarded as representative of the economic reform phase, has been 3.4 percent
per annum, which is more than two percentage points below the 5.5 percent
registered in the period 1950-80; the inclusion of 2009 and 2010 in the calcula-
tions would lower the first of these figures to 3.2 percent. The picture afforded by
Figure 5.10 is even more telling. This figure compares GDP growth per worker
for the different Latin American countries in these two periods, which provides a
good approximation of trends in average labor productivity. 18 This comparison is
much more informative than a comparison of per capita GDP, since demogra-
phic trends tended to depress that indicator during the period of state-led
industrialization, due to the rising demographic dependency ratio in many
countries in the 1950s and 1960s, whereas the "demographic dividend" (i.e., a
4.5%
,0... •Chile
•Panama
0
• •Costa Rica
~ 3.5%
•Guatemala
Q) • • Bolivia Brazil
Q,.....). Colombia M· •exico
~... 2.5%
.~.. •Nicaragua •Honduras •Ecuador
Q) •Venezuela Paraguay•
Q.
0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 4.5%
c0. 1.5% Average growth of GOP per worker 1950-1980
.(...!).
0
~.t: 0.5%
01
Q)
g> -0.5%
£
-1.5%
0.0%
Figure 5.10. Labor productivity, 1990-2010 vs 1950-80
Source: GOP according to ECLAC historical series. The data on economically active population for 1950-80
according to CELADE/ECLAC, for 1990-2010 according to ILO.
18 Strictly speaking, this measurement estimates productivity unadjusted for changes in open
unemployment rates, but it captures the extent of labor-force utilization more accurately.
The words you are searching are inside this book. To get more targeted content, please make full-text search by clicking here.
017_The_Economic_DEVELOPMENT_(book4you.org)_165
Discover the best professional documents and content resources in AnyFlip Document Base.
Search