Structures, Endowments, and Institutions in the Economic History of Latin America

Author(s): John H. Coatsworth

Source: Latin American Research Review, Vol. 40, No. 3 (2005), pp. 126-144
Published by: The Latin American Studies Association
Stable URL: http://www.jstor.org/stable/3662825 .
Accessed: 10/01/2015 17:50

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

The Latin American Studies Association is collaborating with JSTOR to digitize, preserve and extend access to
Latin American Research Review.

http://www.jstor.org

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES, ENDOWMENTS, AND
INSTITUTIONS IN THE ECONOMIC
HISTORY OF LATIN AMERICA

JohnH. Coatsworth
HarvardUniversity

The economic history of Latin America has become more voluminous,

complex, and fascinating in the past decade.1 The new work has already
provoked noteworthy commentaries; one could even write an histori-
ography of the historiography.2 The purpose of this essay is to comment
on (and applaud) the re-emergence of political economy in the economic
history of the region. By this I mean the renewal of interest in the Big
Questions that inspired the structuralists, "cepalinos," Marxists,
dependentistas, and modernizationists of the post-World War II gen-
erations. Economic historians are again worrying about the long, long
run, about the connections between social stratification, political power,
and economic strategy, and about the relative impact of structures, en-
dowments, and institutions on economic growth and development.
This essay first reviews what economic historians take as their key
dependent variables, that is, the productivity of economies and the wel-
fare of the people who make them work. We know much more now than
we did only a decade or so ago about the evolution of gross domestic
product (GDP) per capita and changes in living standards over the past
several centuries. Second, it touches on the subject that Joseph Love ad-
dresses at greater length in his contribution to this issue, namely, the rise

1. Four collections of varying scope provide a representative sample: Stephen Haber,

ed., How Latin America Fell Behind: Essays on the EconomicHistories of Brazil and Mexico,
1800-1914 (Stanford: Stanford University Press, 1997); John H. Coatsworth and Alan M.
Taylor, eds., Latin America and the WorldEconomy Since 1800 (Cambridge: Harvard Uni-
versity Press, 1998); Rosemary Thorp, ed., An EconomicHistory of Twentieth-CenturyLatin
America, 3 vols. (New York: Oxford University Press, 2000); and Victor Bulmer-Thomas,
John H. Coatsworth, and Roberto Cortes Conde, eds., The CambridgeEconomicHistory of
Latin America (New York: Cambridge University Press, forthcoming 2006).
2. On this topic, see Paul Gootenberg's penetrating review essay in a recent issue of
this journal: "Between a Rock and a Softer Place: Reflections on Some Recent Economic
History of Latin America," LARR 39, no. 2 (June 2004): 239-57.

Latin American ResearchReview, Vol. 40, No. 3, October 2005

? 2005 by the University of Texas Press, P.O. Box 7819, Austin, TX 78713-7819

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 127

and fall of structuralism and related macrohistorical approaches to un-

derstanding the determinants of Latin America's relative economic back-
wardness. Third, it takes a look at some of the most interesting recent
efforts to restart discussion and debate on the same macro-level issues
that preoccupied the structuralists and their offshoots and critics. Finally,
as is customary, the concluding section brews up a stew of unanswered
questions, well-intentioned speculation, and gratuitous advice.

Over the past decade or so, new efforts to estimate long-term trends
in productivity and welfare have managed to refine our knowledge of
Latin America's economic history in ways that now make it possible to
ask old questions with greater precision, and even to raise altogether
new questions. Productivity is generally measured as GDP per capita.
Though reliable data are still lacking, and GDP estimates for the nine-
teenth century and earlier will always contain substantial margins of
error, the independent guesswork of many scholars is converging to-
wards consensus on some, though not all, of the main trends over the
past three centuries.
The most ambitious effort to produce estimates of GDP per capita
that are comparable across time and between countries and regions is
that of Angus Maddison, who began publishing historical estimates of
GDP per capita for all world regions in the late 1980s.3He has continued
to adjust and refine his efforts up to at least 2003.4Maddison's estimates,
which go back as far as the year 1 AD, are calculated in constant "inter-
national dollars" of 1990. By converting all his estimates to the same
currency units and adjusting them to eliminate the effects of differences
in price levels, Maddison aimed at providing PPP ("purchasing power
parity") estimates that can then be compared across time and space.
Though his figures are useful as indicators, the pre-twentieth-century
estimates are admittedly quite rough, and not always well documented
or explained.5
Despite methodological differences and continuing debates over par-
ticular cases, however, the Maddison estimates appear to embody an

3. See Angus Maddison, "Explaining the Economic Performance of Nations," in Con-

vergenceof Productivity:Cross-NationalStudies and Historical Evidence,ed. William Baumol,
Richard Nelson, and Edward Wolff (Oxford: Oxford University Press, 1994), 20-61.
4. Angus Maddison, The WorldEconomy:Historical Statistics (Paris: OECD, 2003), 113-
50. A number of the historical estimates of GDP and GDP per capita in this source differ
from estimates Maddison published earlier.
5. Maddison's figures for Mexico, for example, include GDP per capita estimates for
1820, 1850, 1870, and 1877. His estimates are based on figures published by Mexico's
Instituto Nacional de Estadistica, Geografia e Informatica (INEGI) in 1985 for 1800,1845,
1860, and 1877, which Maddison extrapolated between these dates to produce estimates
for the years he needed. INEGI's estimates were actually taken from those I published

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
128 Latin American ResearchReview

emerging consensus on long term trends (see table 1). It is now gener-
ally accepted that the areas of Latin America under effective Spanish or
Portuguese control probably enjoyed per capita incomes on a par with
Western Europe and at least equal to the British colonies that became
the United States well into the eighteenth century. Maddison, for ex-
ample, estimates that in 1700 the thirteen British colonies in North
America had a per capita GDP well below the Caribbean and Mexico
and barely equal to the average for Latin America. Others have put
Cuba's GDP per capita higher than that of the United States until the
1830s, though Mexico had already fallen well behind by 1800.6 It also
seems well established that during the half century after the outbreak of
the independence wars in 1810, little if any economic growth occurred
in most of Latin America. The pampas region of Argentina (cattle hides
and wool) and copper-rich Chile probably began growing earlier than
Brazil and the other former Spanish colonies; Mexican GDP per capita
actually fell substantially after 1810.7
The larger Latin American economies and many smaller ones began to
achieve sustained increases in GDP per capita starting at some point in
the late nineteenth or early twentieth century. Since 1870, according to
Maddison, the eight major economies of Latin America have grown at a
rate roughly equal to that of the United States overall, though faster be-
fore 1930 than afterwards. In fact, until World War II, the big eight econo-
mies grew faster than the average of all the advanced countries that later
formed the Organisation for Economic Co-operation and Development
(OECD). Convergence ended with the era of import-substitution

in 1978, which INEGI converted from 1970 dollars to pesos at the 1970 rate of exchange of
26.5 pesos to the dollar. Maddison divided each of INEGI's figures by exactly 2.5 to con-
vert them from 1970 pesos to 1990 dollars. Maddison's Mexican estimates are found in
Maddison, The World Economy, 191. The original estimates were published in John H.
Coatsworth, "Obstacles to Economic Growth in Nineteenth-Century Mexico," American
Historical Review 83, no. 1 (February 1978): 82; the INEGI figures, which cite this source
are in Estadisticashist6ricasde Mexico, 2 vols (Mexico, 1985), 1:300, 311 (table 9.1).
6. For Cuba, see Pedro Fraile Balbin, Richard J. Salvucci, and Linda K. Salvucci, "El caso
cubano: Exportaci6n e independencia" in Laindependencia americana:Consecuenciaseconomicas,
ed. Leandro Prados de la Escosura and Samuel Amaral (Madrid:Alianza Universidad, 1993),
part II, chap. 3. See also John H. Coatsworth, "Economic and Institutional Trajectoriesin
Nineteenth-Century Latin America," in LatinAmericaand the WorldEconomySince 1800, ed-
ited with Alan M. Taylor (Cambridge: Harvard University Press, 1998), 23-54.
7. Anecdotal evidence suggests that Mexican GDP per capita fell sharply during the
independence wars. Partial recoveries occurred in brief periods and certain regions. See,
for example, Margaret Chowning's useful study of Michoacan that argues for a recov-
ery in agriculture (the sector least affected by Mexico's turmoil) from the late 1830s to
1845 in Wealthand Power in Provincial Mexico: Michoacanfrom the Late Colony to the Revo-
lution (Stanford, CA: Stanford University Press, 1999). However, there is no credible
evidence in the literature to suggest that GDP per capita for the country as a whole
returned to late colonial levels until the early Porfiriato.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 129

Table1 GDP Per Capita (In InternationalDollars of 1990)

1500 1600 1700 1820 1870 1900 2000
Mexico
(Maddison) 425 454 568 759 674 1,366 7,218
Mexico
(Coatsworth) c. 550 755 755 566 642 1,435
Other Latin America 410 431 502 663 683 5,508
(excluding
the Caribbean)
Brazil 400 428 459 646 713 678 5,556
Caribbean countries 400 430 650 636 549 880 5,634
Latin America 416 438 527 692 681 1,110 5,838
USA 400 400 527 1,231 2,445 4,091 28,129
Sources: See Angus Maddison, The World Economy: Historical Statistics (Paris: OECD,
2003), 113-50, 262; John H. Coatsworth, "Mexico" in The Oxford Encyclopediaof Economic
History, ed. Joel Mokyr (New York: Oxford University Press, 2003), 501-07.

industrialization (ISI). Inward-oriented growth strategies worked bet-

ter in some places than others, but did poorly overall. Unfortunately,
the collapse of ISI has not yet produced a successful alternative. In the
last quarter of the twentieth century, Latin American per capita GDP
growth fell to half that of the United States (0.91 percent per year for
Latin America versus 1.86 for the United States in 1973-2001).8
Data on living standards show similar trends: Stagnation until the
late nineteenth century and slow but steady improvement thereafter. In
most of Latin America, for example, life expectancy at birth (30-35 years)
and infant mortality rates (near 300 per 1000 live births) stood at levels
comparable to the Roman empire until 1900 or so; literacy rates in most
countries had not yet reached 30 percent of the adult population.9 Over
the course of the twentieth century, life expectancy has risen to more
than 70 years and infant mortality has fallen by an order of magnitude,

8. Maddison, The World Economy, 263. See also the discussion in the editors' "Intro-
duction" to Gerardo Della Paolera and Alan M. Taylor, eds., A New EconomicHistory of
Argentina (Cambridge, UK: Cambridge University Press, 2003), 1-18. The experience of
individual countries varied, of course. Argentina reaped spectacular gains up to 1913
and grew more slowly than the rest of Latin America after 1930. Mexico's productivity
gains of the pre-1930 era were interrupted by the revolution of 1910, but the Mexican
economy grew faster than most during the ISI era.
9. The Mexican case is representative. See INEGI, Estadisticashist6ricasde Mexico, 2 vols.
(Mexico: INEGI, 1985). Countries and regions of predominantly European settlement with

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
130 Latin American ResearchReview

to 30 or less per 1000 in most countries. New work on the height of

populations going back to the pre-Hispanic era has begun to shed light
on fluctuations in the biological standard of living.10While differences
are apparent across regions and time periods, sustained improvements
are largely a twentieth-century phenomenon. This pattern differs from
that of earlier industrializations in Britain, the United States, and West-
ern Europe where productivity rose rapidly during the nineteenth-
century industrial revolution while living standards fell or stagnated
and did not catch up until decades later.1lIn Latin America, productiv-
ity and welfare indicators rose together in the twentieth century.
Though living standards rose, inequality also increased. Moderniza-
tion appears to have produced a massive new concentration of land
ownership provoked, inter alia, by railroads that brought opportunities
for commercial exploitation to once isolated regions; technological change
(especially in sugar) that created economies of scale; the rapid develop-
ment of large banana plantations in the tropics; and the sale of public
lands in large blocks to land and survey companies and well-connected
entrepreneurs.12 The new concentration of land ownership provoked
violence in the densely populated indigenous provinces of central and

smaller populations of indigenous or African descent tended to raise literacy and achieve
improvements in health indicators more rapidly than others. For literacy rates, see Elisa
Mariscal and Kenneth L. Sokoloff, "Schooling, Suffrage, and the Persistence of Inequality
in the Americas, 1800-1945," in Political Institutions and EconomicGrowthin Latin America,
ed. Stephen Haber (Stanford: Hoover Institution Press, 2000), 165-75.
10. For an excellent recent example of this work, which makes use of an unusually
large data set, see Adolfo Meisel R. and Margarita Vega A., "La estatura de los
colombianos: Un ensayo de antropometria hist6rica, 1910-2002 (Documentos de Trabajo
sobre Economia Regional 45, Centro de Estudios Regionales, Banco de la Republica,
Cartagena, May 2004). See also Moramay L6pez Alonso and Radl Porras Condey, "The
Ups and Downs of Mexican Economic Growth: The Biological Standard of Living and
Inequality, 1870-1950," Economicsand Human Biology 1 (2003): 169-86. The Mexican data
show a decline in heights from the 1870s to the 1920s with improvements thereafter.
11. On living standards during industrial revolutions, see John Komlos, "Shrinking in
a Growing Economy: The Mystery of Physical Stature during the Industrial Revolu-
tion," Journalof EconomicHistory 58, no. 3 (September 1998): 779-802, and Roderick Floud
and Richard Steckel, eds., Health and Welfareduring Industrialization (Chicago: Univer-
sity of Chicago Press, 1997).
12. On the impact of railroads on landownership in Mexico, see John H. Coatsworth,
"Railroads, Landholding and Agrarian Protest in the Early Porfiriato," Hispanic Ameri-
can Historical Review 54, no. 1 (February 1974): 48-71. For the Cuban case, and an inter-
esting reinterpretation of the consequences of concentration in sugar, see Alan Dye, Cuban
Sugar in the Age of Mass Production: Technologyand the Economics of the Sugar Central,
1899-1929 (Stanford: Stanford University Press, 1998); on the vast size of banana planta-
tions, see the classic work by Charles Morrow Wilson, Empire in Green and Gold: The
Story of the American Banana Trade(New York: H. Holt and Co., 1947). On the surveys
and sales of public lands, see Robert H. Holden, Mexico and the Survey of Public Lands:
The Management of Modernization (DeKalb, IL: Northern Illinois University Press, 1994).

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 131

southernMexicoand laterin partsof the Andean highlands.13In Argen-

tina, land ownership tended toward concentrationmuch earlier,as the
governmentrewardedfriendsand Indianfighterswith vast tractsstart-
ing soon after independence and lasting through the "Conquestof the
Desert"in 1879. The effect of these giveaways on the concentrationof
wealthand incomewas limited,however,becauseland values stayed low
until railroadsand refrigeratorships created opportunitiesfor profit-
able commercialexploitationon a large scale.14The trend toward con-
centrationof wealth in the countrysidecoincidedin many countrieswith
risingrewardsfor owners of capital,land, and skillsrelativeto the wages
of unskilled majorities.Immigrationfrom southern Europe to Argen-
tina, Brazil,Uruguay and elsewhere tended to increasethe downward
pressure on wages.l5Unlike other modernizing regions, most of Latin
Americadid not experiencea reversetrend toward great equality later
in the developmentprocess.In the post-WorldWarIIera, two powerful
trends worked against equality.First, the reigning model of state-led
import-substitutionindustrializationtended to increasewage inequal-
ity both by inflating wages in protected industries and by providing
relativelyhigh salariesand benefitsto stateemployees.Second,the Cold
Warpolitical and economic alliancethat the dominantclasses and gov-
erning elites of LatinAmericaforged with the United States gave Latin
Americaa half centuryof governmentsdecidedly moreconservativeon
social issues than most of the region'spopulation.
High levels of inequality had at least two importantconsequences
for most Latin American countries. First, inequality slowed improve-
ments in living standards.Latin American welfare indicators did im-
prove over the course of the twentieth century, but inequality
compounded by high population growth rates made progress slower

13. For an attempt at quantification, see John H. Coatsworth, "Patterns of Rural Re-
bellion in Latin America: Mexico in Comparative Perspective," in Riot, Rebellion, and
Revolution: Rural Social Conflict in Mexico, ed. Friedrich Katz (Princeton: Princeton Uni-
versity Press, 1988), 21-62.
14. On Argentina, see the essay by Tulio Halperin Donghi, "Argentina: Liberalism in a
CountryBornLiberal,"in GuidingtheInvisibleHand:EconomicLiberalism
andtheStatein
Latin American History, ed. Joseph L. Love and Nils Jacobsen (New York: Praeger, 1988),
99-116. See also, Lyman Johnson, "The Frontier as an Arena of Social and Economic
Change: Wealth Distribution in Nineteenth-Century Buenos Aires Province" (Unpub.
paper, n.d.).
15. In Argentina, real wages actually rose over this period, but Alan M. Taylor argues
that without immigration, wages would have been 25 percent higher on the eve of World
War I in "Peopling the Pampas: On the Impact of Mass Migration to the River Plate,
1870-1914," Explorations in EconomicHistory 34, no. 1 (1997): 100-23; see also Jeffrey G.
Williamson, "Winners and Losers over Two Centuries of Globalization" (National Bu-
reau of Economic Research, Working paper 9161 Cambridge, MA: September 2002), 17-
23.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
132 Latin American ResearchReview

than elsewhere. Second, inequality contributed to keeping poverty rates

exceptionally high throughout much of the region into the twenty-first
century. Latin America remains today the most unequal region in the
world.16

Much of what we know now about these historical trends, particu-

larly in quantitative terms, is knowledge that originated in research that
began in the 1950s. Moreover, the issues raised by the postwar genera-
tion of structuralist economists, and their dependentista offspring in the
1960s and 1970s, have now returned to enrich, if sometimes also to frus-
trate, scholarly debate at the dawn of this new century. As Love demon-
strates, the intellectual history of economic debate in the post-World
War II era left a vast and varied landscape of great peaks and valleys,
and not a few swampy areas ready to trap the unwary traveler. We now
know three important things about the structuralists, including particu-
larly the cepalinos (those who worked for, or more broadly those who
were trained by or generally agreed with CEPAL, the UN Economic
Commission for Latin America) that have not received adequate atten-
tion in the historiography. The first is that, in addition to their theoreti-
cal work and their essays in general economic history, the cepalinos
created and implemented ambitious programs of economic research and
data collection in one country after another. In several countries, the
first serious efforts to construct national accounts and develop plausible
estimates of past trends in GDP were carried out by CEPALor by CEPAL-
trained economists.17 It is simply untrue that the structuralists and
dependentistas did not do serious quantitative or qualitative empirical
research; often, they were the first economic historians to do so.18

16. For a survey of the twentieth century, see Rosemary Thorp, Progress, Poverty and
Exclusion:An EconomicHistory of Latin America in the TwentiethCentury (Baltimore: Johns
Hopkins University Press, 1998), chap. 13.
17. On Chile for example, see United Nations Economic Commission for Latin America
(ECLA or in Spanish CEPAL),Antecedentessobreel desarrolloecon6micode la economiachilena
1925-52 (Santiago, Chile: Pacifico, 1954); see also the two volumes on Argentina pub-
lished in 1959 as part of the larger project of the entitled Analisis y proyeccionesdel desarrollo
econ6mico,Part V. El desarrolloecon6micode la Argentina, 2 vols. (Mexico: United Nations
Department of Economic and Social Affairs, 1959).
18. Inexplicably, Stephen Haber's introductory chapter to How LatinAmericaFell Behind
(cited above) suggests that structuralists and dependentistas were guilty of neglect and
even contempt for empirical research. In this case, it was Haber who got the facts wrong.
Even more sweeping, and anachronistic, was Douglass C. North's identification of "Latin
American studies" with "a long tradition... [of] dependency explanations of the region's
lagged growth" and the bizarre suggestion that integrating "history, economics, and poli-
tics . . . has not been part of ... the literature on Latin American development ... " in
North, "Concluding Remarks: The Emerging New Economic History of Latin America,"
in Haber, ed., Political Institutions and EconomicGrowthin Latin America,273.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 133

Second, we now know, too, that CEPAL's endorsement of import-

substitution industrialization based on high levels of tariff protection
had been the dominant economic policy strategy in every major Latin
American economy for at least the preceding half century. Indeed,
Latin America was the most protectionist region in the entire world for
as far back as the data go, that is, the mid-nineteenth century. Mexico,
Argentina, Brazil, Peru, and Chile all adopted explicit policies of indus-
trial protection well before World War I. In the 1930s, when conven-
tional accounts suggest that the region turned inward, Latin America
actually lost its position as the most protectionist region in the world, as
other regions adopted high tariff strategies and caught up. What changed
in Latin America in the 1930s, but then only slowly, was the use of non-
tariff barriers to complement the high tariff regimes that had already
been in place for a century-and the beginnings of state-led industrial
promotionin addition to protection.19
Third, we now also know that the United States supported, promoted
and pressured for the adoption of ISI policies throughout the less-devel-
oped world in the 1950s. It is true that the United States opposed the
creation of CEPALand did its best to discredit the agency and its advice,
but the U.S. objections to CEPALhad nothing to do with ISI per se. They
focused on CEPAL'sadvocacy of state planning and regulation. The United
States feared that CEPAL-influenced economic policies would lead to the
creation of state enterprises or subsidies to domestic companies, either of
which would close off opportunities for U.S. multinational businesses
that were seeking to jump over tariff walls and build branch plants pro-
ducing for domestic markets in the larger Latin American countries.20
The key link between the structuralists and the later dependency
school writers-apart from the fact that they were often the same
people-was the preoccupation of both with institutions as key deter-
minants of economic success or failure. For Raul Prebisch, the key to
understanding the alleged deterioration in terms of trade, which he
thought affected most of Latin America and the developing world, lay
not in the market place but in the institutional arena. Terms of trade
deteriorated for primary product producers because union contracts in
the manufacturing industries of the developed world made it difficult
to lower wages during recessions and oligopolistic industries effectively
colluded to reduce production rather than lower prices when demand

19. See John H. Coatsworth and Jeffrey G. Williamson, "Always Protectionist? Latin
American Tariffs from Independence to Great Depression," Journalof LatinAmericanStud-
ies 36, no. 2 (May 2004): 205-32.
20. Sylvia Maxfield and James Nolt, "Protectionism and the Internationalization of
Capital: U.S. Sponsorship of Import Substitution Industrialization in the Philippines,
Turkey and Argentina," International Studies Quarterly 34 (1990): 44-81.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
134 Latin American ResearchReview

fell.21The absence of such effective private institutions in the third world

made it impossible to pursue a similar strategy there. Within Latin
America, as Love points out, the structuralists focused on institutional
"structures," such as the concentration of land ownership. One of the
most important debates erupted over the structuralist claim that
latifundismocaused inflation by limiting the supply of agricultural prod-
ucts. The Chicago School economists effectively rebutted this argument
by showing inflation stemmed not from inelastic supplies of agricul-
tural (or other) products but from an oversupply of money linked to
government deficits.22
The generation of dependency school writers of the 1960s and 1970s
is conventionally assumed to have been interested primarily in the eco-
nomic relations between the developed core of capitalist industrial na-
tions and the underdeveloped periphery. Actually, most of the
dependentistas, such as Fernando Henrique Cardoso and Enzo Falleto
to take the iconic case, devoted almost no time at all to analyzing Latin
America's relations with the developed world.23 Instead, they focused
attention on the impact of trends in the external sector (including export
production and foreign direct capital investment) on the evolution of
domesticclass formation, economic structure, political coalitions and in-
stitutions. Much of what they discovered and wrote about on these is-
sues is now accepted as commonplace and uncontroversial. Much less
durable were the dependentista models that focused directly on exter-
nal economic relationships. Prebisch's hypothesis on the terms of trade
was probably wrong, though still debated.24 Models of "unequal ex-
change" or surplus extraction and transfer from the periphery to the
core could not be verified empirically.25

21. See United Nations Economic Commission for Latin America, The EconomicDevel-
opment of Latin America and its Principal Problems(Lake Success, N.Y.: CEPAL, 1950); see
also Joseph Love's essay in this issue.
22. See Love's discussion of the structuralists in this issue. For a clear expression of
the view, now universally accepted, that money supply determines the rate of inflation
in any structural environment, see Arnold Harberger, "El problema de la inflaci6n en
America Latina," Boletin Mensual of the Centro de Estudios Monetarios (Buenos Aires,
Argentina), June 1966: 253-69.
23. Fernando Henrique Cardoso and Enzo Faletto, Dependencyand Developmentin Latin
America (Berkeley: University of California Press, 1979).
24. On the terms of trade debate, see Christopher Blatman, Jason Hwang, and Jeffrey
G. Williamson, "The Terms of Trade and Economic Growth in the Periphery, 1870-1983"
(Working Paper 9940 National Bureau of Economic Research, Cambridge, MA, August
2003).
25. See for example, the late Andre Gunder Frank's influential Capitalismand Underde-
velopment in Latin America:Historical Studies of Chile and Brazil (New York: Monthly Re-
view Press, 1967); less well known outside Latin America was the work of Ruy Mauro
Marini, Dialectica de la dependencia(Mexico: Ediciones Era, 1973).

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 135

Most important of all, the consensus among economists and economic

historians of Latin America on the relationship between external ties
and economic performance has now shifted 180 degrees from half a cen-
tury ago. The structuralist and dependentista generations believed that
Latin America's external economic ties to the developed world would
promote economic growth only if carefully regulated by a powerful state.
Many, like Celso Furtado, argued that past episodes of export booms
had depleted soils or extracted minerals and ended with "involutions,"
economic downturns that left entire regions and economies worse off
than before. Most believed that vigorous state intervention could ma-
nipulate external trade and capital flows to promote domestic industry
and escape this historic pattern, but some writers were more pessimis-
tic.26Economic historical research in the past two decades, some of it
based on empirical work initiated by the dependentistas themselves,
has now produced a contrary consensus: In the long run, countries open
to external trade and capital flows have tended to grow faster and reach
higher levels of GDP per capita sooner than others. This is true for as far
back as the late colonial era.27
Since the structuralist and dependentista scholars apparently got the
growth issue wrong, succeeding scholars have tended to overlook what
they got right. Three of the key insights in this literature are now re-
emerging, so it is particularly relevant to point them out. First, the
dependentistas insisted that the less-developed world could not suc-
ceed economically by following the path trod by the developed nations.
To some extent, this insistence arose as a response to the equally insis-
tent modernizationist view that the Anglo-American model, understood
as demonstrating the efficacy of open markets and free trade, should be
adopted by Latin America. The counter-argument was all too easy to
make: Latin America was different, and faced a different international
environment in the late twentieth century than Britain and the United
States did in the early nineteenth century.
Second, they claimed, accurately it appears, that external dependence
exacerbated economic inequality in the less-developed world. Fernando
Henrique Cardoso argued, for example, that external dependence did
not make economic growth impossible, but merely immoral.28Of course,
long before the onset of export-led economic growth, the subjugation of

26. See the lucid discussion of Furtado in Love's essay in this issue.
27. See Coatsworth, "Trajectories." For an econometric test for the twentieth century,
see Alan M. Taylor, "On the Costs of Inward-Looking Development: Price Distortions,
Growth, and Divergence in Latin America," The Journal of Economic History 58, no. 1.
(March 1998): 1-28.
28. See Fernando Henrique Cardoso, "Industrialization, Dependency and Power in
Latin America," BerkeleyJournalof Sociology 17 (1972): 79-95.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
136 Latin American ResearchReview

Native Americans and the enslavement and forced migration of Afri-

cans had already embedded inequality in the fabric of many Latin Ameri-
can societies. The relative weight to be given to the historic inequality in
legal status and class structure associated with caste and slave systems,
as opposed to the inequality in income and wealth exacerbated by onset
of modernization is still being debated (see below). Third, the structur-
alist and dependentista emphasis on issues of political economy and
institutional development also has a newly contemporary ring. Few of
the cepalinos then or now would be surprised to learn that economic
historians, along with economists and political scientists, are now tend-
ing to the view that inequality leads to the creation of institutions that
inhibit economic performance.
The military regimes that closed off political debate throughout Latin
America between 1963 and the 1980s left a permanent mark on the intel-
lectual history of the region. Some of the military regimes (for example,
in Argentina and Brazil) espoused a kind of primitive nationalism that
proved to be consistent with structuralist and dependentista arguments
about the need for state-led management of both the domestic economy
and external economic relations. Few followed the Chilean regime's
decision to embrace open markets and free trade. Though an odd scat-
ter of dependentistas turned up working for such military rulers, the
main effect of military rule was to put an end to grand theories as well
as utopian visions. It was not just that these regimes overtly stigmatized
and punished such thinking; more profound was the widespread belief
that such theories and visions had failed.
In economic history, the turn from macrohistorical questions to
microeconomic issues coincided with this general trend. The coincidence
was obscured by the persistence of competition between the still contro-
versial "neoclassical" approach to economics and older institutionalist
and Marxist traditions. The arrival in Latin America of the "New Eco-
nomic History," which first swept the United States in the 1960s, was
probably delayed both by local resistance and by the relative decline in
academic research on Latin America in the United States that occurred
in the 1970s and 1980s. These caveats aside, the focus of the new eco-
nomic historians on questions derived mainly from microeconomic
theory was consistent with trends in other disciplines. Moreover, the
emergence of the Washington Consensus29 and the collapse of socialism
in Eastern Europe contributed to the view that, even if history had not

29. The term "Washington Consensus" was coined by John Williamson in a volume of
papers he edited entitled LatinAmericanAdjustment:How Much Has Happened?(Washing-
ton, D.C.: Institute for International Economics, 1990). The Washington Consensus in
Williamson's definition included better fiscal management, deregulation of product, fac-
tor, and currency markets, privatization of public assets, and more effective governance.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 137

quite ended, most of the Big Questions had now been settled. The re-
search of this era looked at the impact of technological change (espe-
cially railroads),30industrial and business history,31and the design of
policies and institutions in key sectors of the region's larger economies.32
This work contributed significantly to the economic historiography of
Latin America. It also makes it possible to return to the Big Questions
with a much more solid empirical and analytical foundation.

In the twenty-five years of the Washington Consensus, the Latin

American economies have experienced their worst quarter century since
the catastrophic second quarter of the nineteenth century. Initially, the
collapse of growth rates was attributed to the effects of the financial
crisis precipitated by the Mexican devaluation and de facto default in
August 1982. By the mid-1980s, however, most of the region had begun
abandoning the failed ISI strategies of the previous half century and
was embracing market-friendly reforms, including greater openness to
external trade and investment. While growth rates did recover some-
what in the 1990s, they fell again in the recession that greeted the twenty-
first century. In a number of cases, growth remained anemic or proved
short-lived despite major economic policy reforms. The empirical evi-
dence linking growth rates to Washington Consensus structural reforms
is weak (except for openness to trade, which is associated with faster
growth).33Economists have generally argued, nonetheless, that a major
determinant of the crises and difficulties in the 1990s and early 2000s
was failure to implement the reforms fully, especially those related to
fiscal restraint.34Most also take the view that while economic reforms
may constitute a necessary condition for success, they are not sufficient

30. John H. Coatsworth, GrowthAgainst Development:The EconomicImpact of Railroads

in Porfirian Mexico (DeKalb: Northern Illinois University Press, 1981); William R.
Summerhill, OrderAgainst Progress:Government,Foreign Investment, and Railroadsin Bra-
zil, 1854-1913 (Stanford: Stanford University Press, 2003).
31. See, for example, Richard J. Salvucci, Textilesand Capitalismin Mexico: An Economic
History of the Obrajes, 1539-1840 (Princeton, N.J.: Princeton University Press, 1987);
Stephen Haber, Industrialization and Underdevelopment (Stanford: Stanford University
Press, 1989).
32. See, for example, Stephen Haber, "The Efficiency Consequences of Institutional
Change: Financial Market Regulation and Productivity Growth in Brazil, 1866-1934," in
Latin America and the World Economy Since 1800, ed. John H. Coatsworth and Alan M.
Taylor (Cambridge, MA: Harvard University Press, 1996), 275-322.
33. Eduardo Lora and Ugo Panizza, "Structural Reforms in Latin America Under Scru-
tiny" (Research Department Working paper Washington, DC, Inter-American Develop-
ment Bank, 2002).
34. See, for example, Arminio Fraga's balanced summary in "Latin America Since the
1990s: Rising from the Sickbed," Journal of EconomicPerspectives 18, no. 2 (Spring 2004):
1-18.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
138 Latin American ResearchReview

to guarantee faster growth. Some have pointed to geographic obstacles

to economic growth,35but in recent years discussions have focused in-
creasingly on institutional deficiencies-inadequate protection of prop-
erty rights, inefficient and corrupt bureaucracies and courts, and multiple
public sector failures evidenced by widespread tax evasion, procure-
ment fraud, crumbling infrastructure, inadequate regulatory systems,
lagging education and health services, and the like.36
Institutions have histories. Most institutional histories contain two
key elements: politics and path dependence. Institutions are created and
sustained by private groups and public entities in the political arena.
They tend to come in mutually reinforcing and complementary clusters
that persist in the absence of external shocks or endogenous threats.
Conditions of rough but durable equilibrium create the path-dependent
trajectories that open or foreclose opportunities for long-run economic
growth. The political economy of economic failure has thus re-emerged
as a central preoccupation of economic historians of Latin America.
One of the most influential new works in this re-emerging field was
the seminal essay by Stanley Engerman and Kenneth Sokoloff, first pub-
lished in a volume of essays edited by Stephen Haber.37To summarize
brutally, Engerman and Sokoloff argued that factor endowments (the
prevailing quantity, quality, and relative scarcities of land, labor, and
capital) in Latin America in the colonial era gave rise to high levels of
concentration of wealth, particularly land ownership, in contrast to the
relatively egalitarian distribution of wealth in the northern colonies of
British North America. Excessive concentration of wealth led to concen-
tration of political power in the hands of narrow elites and thus to the
creation of institutions that served elite interests and failed to protect
the property rights of most citizens. In the mid-Atlantic and northern
United States, by contrast, factor endowments favored family farms

35. For an early discussion, see, J. L. Gallup, Jeffrey Sachs, and A. Mellinger, "Geogra-
phy and Economic Development," Annual World Bank Conferenceon Development Eco-
nomics 1998 (Washington, D.C.: World Bank, 1999); for an application to Latin America,
see Gerardo Esquivel, "Geografia y desarrollo econ6mico en Mexico" (Research Net-
work Working Paper No. R-389 Inter-American Development Bank, Washington, D.C.,
April 2000). Geographic obstacles include absence of natural resources, difficult climatic
conditions, distance from navigable waterways, and the like.
36. See, for example, Harold L. Cole, Lee E. Ohanian, Alvaro Riascos, and James A.
Schmitz, Jr., "Latin America in the Rearview Mirror" (National Bureau of Economic Re-
search Working Paper 11008 Cambridge, MA, December 2004), which argues that Latin
America's poor growth performance is the result of barriers to both external and do-
mestic competition such as protectionism and barriers to entry that privilege existing
producers to a far greater extent than in other world regions.
37. Stanley Engerman and Kenneth Sokoloff, "Factor Endowments, Institutions, and
Differential Growth Paths among New World Economies," in Haber, ed., How America
Fell Behind, 260-304.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 139

rather than large estates. Greater equality in wealth distribution diffused

political power more broadly and made it possible to create institutions
that protected citizens and their property from arbitrary treatment by
government. As a result, people in the United States made broad and
continuous use of markets to exchange products and factors, while most
Latin Americans, facing higher risks, did not.
The Engerman and Sokoloff thesis, while plausible, is almost certainly
wrong. Though the institutional framework for colonial economic ac-
tivity did not vary much across the Spanish American colonies, GDP
per capita measured at the end of the colonial era diverged significantly.
The gap between the richest and poorest Spanish colonies was almost
as great as the gap between the richest and poorest regions in the world
in the early nineteenth century. In 1800, for example, Cuba had a per
capita income of $90, compared to roughly $30 for less-developed re-
gions like Brazil and Peru, a ratio of three to one.38Maddison's figures
for the world economy in 1820 show a ratio of four to one between West-
ern Europe and Africa, the richest and poorest regions respectively.39
Moreover, Spanish slave colonies like Cuba, where inequality must have
been greatest, had the highest per capita GDP at that time. Most of the
Spanish colonies were not slave colonies and, however rich or poor, there
is no solid evidence to suggest unusually high levels of concentration of
landownership. Unlike Western Europe, Latin America's colonial elites
did not monopolize land ownership. Throughout Mesoamerica and the
Andes, indigenous villages and villagers occupied most of the arable
lands; European estates clustered in the commercially more profitable
areas near cities and towns and along major trade routes but left vast
areas of the landscape in indigenous hands.40 Even in the frontier and
peripheral areas of the Argentine interior, where huge estates existed on
paper, the value of the land was minimal and contributed little to con-
centrating wealth; the real source of wealth, the cattle, ran wild and be-
longed to no one until long after independence. Thus, landownership
(and wealth more generally) was not more concentrated in Latin America
than in the thirteen British colonies (or industrializing Britain itself).
Moreover, as near as one can tell, inequality was positively correlated
with GDP per capita, both within and between the Spanish colonies in
the Americas, the case in point being the Caribbean slave colonies, which
were enormously productive by any measure of output value. Since the
Engerman and Sokoloff model posited a straightforward linear path
dependence between an initially high concentration of landownership

38. Coatsworth, "Economic and Institutional Trajectories," 29.

39. Maddison, The WorldEconomy,262.
40. The contrast here is to Western Europe where powerful aristocracies excluded most
peasants from owning arable land, even in areas where serfdom had never developed.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
140 Latin American ResearchReview

in the colonial era, bad institutions, and relative backwardness, it can

safely be discarded on empirical grounds.
The origins of the colonial property rights regime, along with slavery,
caste systems, and most other economically relevant institutions cannot
be found in New World factor endowments but in Old World policies and
practices, adapted as needed to New World conditions. It was not New
World factor endowments but Old World conquest and enslavement of
ethnically distinct populations that made for "bad" institutions. But even
if the fundamental institutions that defined economic organization in Latin
America were Iberian in origin, it is still possible that colonial elites domi-
nated the adaptive process and shaped them to suit their own interests.
This did occur, it is true, but only within limits imposed by the colonial
regime itself. For example, colonial elites in Spanish America could not
enslave the indigenous population (after the 1540s). Moreover, the Crown
also made sure that the indigenous population did not depend on the
Creole settler elites for access to land and insisted that the royal courts
protect indigenous lands from usurpation by local magnates. Colonial elites
could not dominate colony-wide governments, repeal taxes and regula-
tions on trade, abolish crown monopolies, or make war or peace. Colonial
elites were first of all colonial,that is, relatively weak in comparison to
European landowning aristocracies, so even if wealth had been highly
concentrated, the Creole elites of Spanish America could not have remade
the institutions that governed economic life just as they pleased.41
The concentration of wealth and elite institutional power that Engerman
and Sokoloff attributed to colonial factor endowments did, in fact, arise in
Latin America but much later, for different reasons, and with contrary re-
sults. They marked the onset of modernizationin LatinAmerica at the end of
the nineteenth and beginning of the twentieth century. What Engerman
and Sokoloff saw as obstacles to economic growth-elite power and eco-
nomic inequality-actually facilitated the region's transition to sustained,
if unstable, economic growth for most of the twentieth century. This rup-
ture with the past took decades to achieve. Creole elites had to settle their
internalfeuds, mobilize resourcesto suppress indigenous peasants and other
competitors for political power, undertake far-reachinglegislative changes
(from abolishing archaic forms of property holding to writing new com-
mercial codes), and attractforeign capital to finance infrastructureand ex-
port production. The result was not stagnation, but growth.
How did Latin America achieve growth, saddled with what Engerman
and Sokoloff condemned as obstacles to broad-based participation in

41. John H. Coatsworth and Gabriel Tortella Casares, "Institutions and Long-Run Eco-
nomic Performance in Mexico and Spain, 1800-2000" (Working Papers on Latin America,
Paper No. 02/03-1 David Rockefeller Center for Latin American Studies, Harvard Uni-
versity, 2002).

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 141

the market? This issue was tackled in a recent work by Stephen Haber
and his collaborators.42"Crony capitalism"-or more awkwardly, "ver-
tical political integration" (VPI), as Haber, Armando Razo, and Noel
Maurer term it-is a system composed of "a rent-seeking coalition made
up of asset holders, a government too weak to establish a despotic state,
and a group that receives rents in exchange for enforcing the contract
between the asset holders and the government."43Rather than "limited
government" with effective institutional constraints on governmental
predation, which would have protected and served all citizens, the best
Mexico could do after a half century of war and instability, was the
Porfirian system of institutionalized cronyism that produced economic
growth by guaranteeing protection to a small elite of politically con-
nected bankers, industrialists, and foreign companies. The 1910 revolu-
tion fractured the Porfirian system and then virtually destroyed it in the
chaotic period of civil warfare (1914-17), but even before the fighting
had ended new deals were being cut.
The VPI model proposed by Haber, Razo, and Maurer provides a
useful framework for addressing a major issue in Latin America's eco-
nomic historiography, that is, how to account for the region's respect-
able rates of economic growth in the twentieth century, despite its
persistent political instability. Extending this analytical framework to
the rest of Latin America would help to refine and deepen our under-
standing of the region's modern political economy. The model is best at
illuminating relationships that turn self-interested deal making into the
kind of credible commitments against expropriation (in its many forms)
that persuade capitalists to make investments. It is less successful, how-
ever, in approaching the macroeconomic issues that confront VPI gov-
ernments with long time horizons. In the Mexican case, for example,
Haber et al. treat the creation of Banamex in the 1880s as a case of in-
sider deal making in which the Mexican government got a stream of
revenue from the bank in exchange for privileges that could not be re-
voked without damaging the government itself. This is true as far as it
goes, but as Haber et al. recognize, the Diaz government was also play-
ing for higher stakes, seeking to use the banks as part of a strategy to
achieve a coherent set of monetary, fiscal, and development objectives.44

42. Stephen Haber, Armando Rozo, and Noel Maurer, The Politics of Property Rights:
Political Instability,CredibleCommitments,and EconomicGrowthin Mexico, 1876-1929 (Cam-
bridge, UK: Cambridge University Press, 2003).
43. Ibid., 29. In the authors' conception, VPI is the institutionalized version of crony
capitalism.
44. On the Porfirian regime's strategy to use the new bank in the service of larger
policy goals, see Thomas Passananti, "Managing Finance and Financiers: The Politics of
Debt, Banking and Money in Porfirian Mexico" (Unpub. Ph.D. dissertation, University
of Chicago, 2001).

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
142 Latin American ResearchReview

VPI governments were not merely predatory, as the model assumes, but
at times farsighted. By the 1930s, many of these regimes even proved
flexible enough to take a temporarily populist turn.
It would be a mistake, however, to reject the hypothesis of a signifi-
cant link between colonial developments and modern economic perfor-
mance. Indeed, the connection between the Iberian colonial regimes and
economic backwardness has held a prominent place in writing about
Latin America since before independence. This conventional idea re-
ceived a decidedly unconventional boost recently in a series of papers
by Daron Acemoglu, Simon Johnson, and James A. Robinson, the first of
which, entitled "The Colonial Origins of Comparative Development: An
Empirical Investigation," appeared in the American EconomicReview in
2001.45Acemoglu, Johnson, and Robinson showed that colonial areas
with high death rates (mainly from tropical diseases) attracted fewer
European settlers and those that did stay served mainly as colonial ad-
ministrators and exploiters of local or imported slave populations. These
areas tend to have less-productive modern economies, while colonies
that attracted high proportions of European settlers tended to do better
over the long run. The authors attributed the modern income gap be-
tween the two sets of colonies, like Engerman and Sokoloff, to differ-
ences in institutions. Colonies ruled by narrow foreign elites were
saddled with institutions that did not protect the property rights of or-
dinary citizens; colonies with predominantly European populations
ended up with good institutions. In contrast to Engerman and Sokoloff,
who argued that inequality in wealth determined institutional develop-
ment, Acemoglu, Robinson, and Johnson argued for the primacy of in-
equalities in legal status and political rights, determined not by factor
endowments but by the structure of colonial rule. A later paper by the
same authors documented what they called "The Reversal of Fortune."46
Areas of the globe that were the most urban and advanced before Euro-
pean colonization (such as Central Mexico and the Peruvian highlands),
they argue, fell furthest behind after conquest. In this case, the causal
mechanism is similar-the presence of dense populations of indigenous
people whom the colonial rulers subjugated and exploited. On this read-
ing, the key source of the institutional deficiencies that have inhibited
growth and development in the twentieth century can be found not in
factor endowments per se but in the political economy of conquest and
enslavement.

45. American EconomicReview, 91, no. 5. (December 2001): 1369-405.

46. Acemoglu, Johnson, and Robinson, "Reversal of Fortune: Geography and Institu-
tions in the Making of the Modern World Income Distribution," QuarterlyJournalof Eco-
nomics, 117 (November 2002): 1231-94.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
 STRUCTURES,ENDOWMENTS,AND INSTITUTIONS 143

This recent work has raised, or perhaps merely raised anew, two major
questions addressed by the structuralist and dependentista historians
and economists two generations ago. The first question is whether the
evidence available will permit a systematic investigation of the precise
mechanisms that link colonial society, politics, and institutions to eco-
nomic performance during and beyond the colonial era. The second
question is how the new political economic equilibrium that emerged in
the late nineteenth century (Haber's "crony" capitalism or vertical po-
litical integration) may have been shaped by the colonial legacy it fi-
nally replaced. Contrary to Engerman and Sokoloff, however, neither
can be answered by looking at colonial land holding. Political and eco-
nomic power in Latin America since pre-Hispanic times has accumu-
lated powerfully only when urban rulers have managed to dominate
the countryside.
The late-nineteenth-century political economic rupture came after
prolonged turmoil and struggle in which the balance shifted decisively
in favor of economic elites, but this shift could only have occurred in an
era of export-led globalization. Haber and his collaborators demonstrated
this connection clearly in attributing Diaz's success in creating his pio-
neering VPI regime in Mexico to "an advantage that was not available
to any previous Mexican president: the ability to integrate with the U.S.
economy."47 This hypothesis will strike some readers as excessively
dependentista, but it is almost certainly right. Without railroads to make
Mexico's resources economically accessible, along with the foreign (and
domestic) investment their construction provoked, the Diaz regime
would have failed. Throughout Latin America in the late nineteenth and
early twentieth centuries, VPI regimes succeeded where predecessors
had failed because of globalization. Thus, I suspect as these questions
continue to be explored, students and scholars will soon find themselves
looking for additional insights in some of the ancient structuralist and
dependentista texts many stopped reading a decade or two ago.

A good diagnosis is at least as important as finding the right medi-

cine. As Latin Americans (and Latin Americanists) search for a way out
of the economic stagnation that the region has experienced over the past
quarter century, critiques of the Washington Consensus and globaliza-
tion will surely intensify. To the extent that such critiques succeed in
persuading voters that the region would be better off with less external
trade and investment, less imported technology, higher inflation, and
more regulation and state ownership, they should probably be seen as
costly distractions at best.

47. Haber et al., Politics of Property Rights, 47.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions
144 Latin American ResearchReview

On the other hand, perhaps the region is witnessing not so much the
collapse of the Washington Consensus as the slow-motion disintegra-
tion of the political economy that emerged at the end of the nineteenth
century. This system of weak institutions (even with big and occasion-
ally repressive governments) and concentrated wealth generated sus-
tained economic growth and gradual improvements in living standards
for most of the twentieth century. It proved flexible enough to incorpo-
rate statist or populist variants, but has suffered in recent years as neo-
liberal policies imposed initially by external conditions and interests have
reduced the capacity of governments to sustain the privileged position
of many protected sectors and groups. Even so, it might have adjusted
and survived had it not been for the transition to democratic rule through-
out much of the region.
The "economic system in which property rights are specified and
enforced as private goods"48 now appears exhausted. As Haber, Razo,
and Maurer put it, this system "requires the creation of rent-seeking
coalitions, is economically inefficient, has negative consequences for the
distribution of income, implies political authoritarianism, and requires
that government be an inefficient provider of public services."49It may
be that Latin America is on the cusp of a new rupture with the past, one
that could eventually lead to more equal societies, stimulate a broader
and more popular capitalism (a la Engerman and Sokoloff), and serve to
create a more efficient, effective, and transparent state apparatus. Some
countries seem already well launched in this direction. In many, the winds
of change are pushing ships of state to the left, a tack that carries great
promise but also grave risks. New systems of political and economic
power seldom emerge from straight lines on a legible map, so scholars
as well as policy makers and economic actors may find it convenient to
not only follow developments with a healthy skepticism, but also
to raise their tolerance for political, social, and even economic experi-
mentation in coming years.

48. Ibid., 41.

49. Ibid., 20.

This content downloaded from 128.235.251.160 on Sat, 10 Jan 2015 17:50:25 PM

All use subject to JSTOR Terms and Conditions