SciELO - Scientific Electronic Library Online

 
 número68Revisitando la política industrial y la industrialización en la América Latina del siglo XXReintroducir el estructuralismo: órdenes de acceso limitado, ISI “extrema” y desarrollo índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google

Compartir


Revista de Estudios Sociales

versión impresa ISSN 0123-885X

rev.estud.soc.  no.68 Bogotá ene./abr. 2019

https://doi.org/10.7440/res68.2019.02 

Dossier

CEPAL and ISI: Reconsidering the Debates, Policies and Outcomes*

CEPAL e ISI: reconsideración de los debates, las políticas y los resultados

CEPAL e ISI: reconsideração dos debates, políticas e resultados

Colin M. Lewis** 

** Ph.D. in Economic History. Professor Emeritus of Latin American Economic History at the London School of Economics & Political Science, United Kingdom. Latest publications: Historical Dictionary of Argentina, (Bernardo Duggan, co-author). Lanham: Scarecrow Press, 2019; Latin America, Economic Imperialism and the State. The Political Economy of the External Connection from Independence to the Present, (Christopher Abel, co-editor). London: Bloomsbury Press, 2015. c.m.lewis@lse.ac.uk


ABSTRACT:

Reappraising cepalina import-substituting industrialisation (ISI) means exploring strategies associated with diagnoses and policy prescriptions, recognising that ideas and processes changed over time, that there was divergence between original diagnoses of development problems and policy interventions designed to resolve them, and that some strategies and outcomes attributed to ECLA were distant from original propositions. Section one locates the ECLA project within a stylised chronology. The second focusses on two sub-periods: (i) proto-cepalismo; (ii) the classic phase of cepalismo. The third re-evaluates the project, challenging contemporary and current opinions. The main findings emphasise continuities between the two sub-periods, arguing that current vilification of the project by the left and the right is/was myopic. Industrial growth was underway in some economies by/before the end of the nineteenth century; industrialisation may have occurred in some countries before 1940; and what had been achieved by the 1990s in parts of the continent was considerable.

KEYWORDS: Development; industry; policy; State intervention

RESUMEN:

Revaluar la industrialización por sustitución de importaciones (ISI) cepalina implica explorar estrategias asociadas con diagnósticos y prescripción de políticas y, a la vez, reconocer que las ideas y los procesos cambiaron con el tiempo, que hubo divergencias entre los diagnósticos originales de los problemas de desarrollo y las intervenciones políticas diseñadas para resolverlos, y que algunas estrategias y resultados atribuidos a la CEPAL se alejaban de sus propuestas originales. La primera sección localiza el proyecto de la CEPAL dentro de una cronología estilizada. La segunda sección se centra en dos subperíodos: (i) el proto-cepalismo y (ii) la fase clásica del cepalismo. La tercera revalúa el proyecto y cuestiona algunas opiniones contemporáneas y actuales. Los hallazgos principales enfatizan las continuidades entre los dos subperíodos, argumentando que el vilipendio actual del proyecto por parte de la izquierda y la derecha ha sido y sigue siendo miope. El crecimiento industrial ya estaba en marcha en algunas economías desde antes del final del siglo XIX; la industrialización pudo haber ocurrido en algunos países antes de 1940; y lo que se alcanzópara la década de losnoventa en algunas partes del continente fue notable.

PALABRAS CLAVE: Desarrollo; industria; intervención estatal; política

RESUMO:

Reavaliar a industrialização por substituição de importações (ISI) cepalina implica explorar estratégias associadas com diagnósticos e prescrição de políticas e, por sua vez, reconhecer que as ideias e os processos mudaram com o tempo, que houve divergências entre os diagnósticos originais dos problemas de desenvolvimento e das intervenções políticas criadas para resolvê-los, e que algumas estratégias e resultados atribuídos à CEPAL se afastavam de suas propostas originais. A primeira seção localiza o projeto da CEPAL dentro de uma cronologia estilizada. A segunda seção se centra em dois subperíodos: (i) o proto-cepalismo e (ii) a fase clássica do cepalismo. A terceira reavalia o projeto e questiona algumas opiniões contemporâneas e atuais. As principais descobertas enfatizam as continuidades entre os dois subperíodos e argumentam que o vilipêndio atual do projeto por parte da esquerda e da direita foi e continua sendo míope. O crescimento industrial já estava em andamento em algumas economias desde antes do final do século XIX; a industrialização pôde ter ocorrido em alguns países antes de 1940, e o que se alcançou para a década de 1990 em algumas partes do continente foi notável.

PALAVRAS-CHAVE: Desenvolvimento; indústria; intervenção estatal; política

Industrialization, with more rational policies for import substitution industry (ISI), was recommended [in R. Prebisch The Economic Development of Latin America and its Principal Problems, New York, UN, 1950]. Under Prebisch’s leadership, the Commission’s key policy thesis was that, unless government took corrective action, the existing form of “spontaneous” ISI would have negative welfare effects. The call for industrialization was not in itself new for Latin Americans, who had found inspiration in the ideas of the German Friedrich List, the Romanian Mihail Manoilescu, and the Argentinean [sic.] Alejandro Bunge, the last a former teacher of Prebisch. The difference was that, inspired by new international insights, Prebisch embraced active government intervention, arguing that industrialization had to be planned, or “programmed”, to use CEPAL’s language. To speak of “programming” was new and challenging in connection with the idea of ISI, as was the need for a regional dimension for such a policy to be efficient. (Rivarola Puntigliano & Appelqvist 2011, 41)

The Great Depression of the 1930s and the world trade shocks triggered by the Second World War dealt a fatal blow to export-led growth […] This opened the way for the emergence of a new development pattern in Latin America that we will refer to as state-led industrialization. This term combines the two main characteristics of this process: an increasing focus on industrialization as a mainstay of development and a considerable expansion of the scope of state action in economic and social affairs. The third characteristic of this pattern was that it was geared towards the domestic market. This orientation has been referred to in ECLAC writing as “inward-looking development” but is more commonly known as “import-substitution industrialization” […] however, import-substitution was not the most salient feature of this pattern over time, nor one that was shared by all countries during the half-century in which this development strategy held sway. (Bértola & Ocampo 2012, 138)

The largest economies in South America had one of the most impressive rates of industrial catch up between the late nineteenth century and the late 1970s (Bénétrix et al. 2012; Williamson, 2006). The largest economies in the region had rapid catch up before 1920, in the 1930s, and then Brazil and Colombia had very rapid and sustained catch up with the global leaders between 1940 and 1980. The 1980s and early 1990s slowed down the region, but by the first decade of the twenty-first century all major economies started to industrialize rapidly again. As such, this rapid process of industrialization is one of the most impressive and important processes in the economic history of the Western World and deserves careful scrutiny. (della Paolera, Durán Amorocho & Musacchio 2015, 2)

Introduction

As the quotations above illustrate, explanations of industrialisation in Latin America differ considerably, and have changed over time. For much of the period addressed in this article, the prevailing orthodoxy was that the processes dated from c.1930. This reflected the view that insertion in the global economic system was inimical to the development of manufacturing in peripheral economies -structural and conjunctural factors associated with the terms of trade, differing elasticities of demand for primary commodities and secondary goods, plus the distinct functioning of factor markets in centre and peripheral/satellite economies was inimical to industrialisation (taken to mean development) in the latter. To over-simplify, concerted state action in favour of industry was the only remedy. Revisionist explanations, in contrast to demand-side constraints identified by structuralists, later refined by dependistas, emphasised supply-side factors associated with growth phases driven by economic engagement with the world economy. These included factor and technology flows -not least entrepreneurship, the thickening of the market and consolidation of market-making institutions, and, subsequently, a particular interpretation of (manufactured) export-led growth in some East Asian economies (Prebisch 1950; Furtado 1959; Hirschman 1963; Cardoso & Faletto 1979; Gereffi & Wyman 1990; Amsden 1989 & 2001; Love 2005; Williamson 2011).1 Reflecting on some of these interpretations and theories, this analysis of the debate about industry in Latin America reappraises ideas about industrial growth and industrialisation in Latin America during the heyday of cepalina import-substituting industrialisation (ISI). It focusses on strategies that were (or came to be) associated with the diagnoses and policy prescription of the UN Economic Commission for Latin America (ECLA, later ECLAC).2 The article also recognises that such ideas -as well as the processes themselves- changed during the time periods considered, and that there was considerable divergence over time and space between original diagnoses of contemporary development problems and evolving policies and strategies designed to resolve those problems. As such, the paper attempts to appraise ideas and strategies, as much as the process of industrialisation per se. In addition, the articles emphasises that some diagnoses, strategies and outcomes attributed to the Commission were rather distant from its original propositions.

Returning to the opening quotations, two simplistic adverse criticisms of ECLA and structuralism are easily made. First, that many advocates of ISI conflated industrialisation and development or mistook industrialisation for development. Secondly, that ISI as applied in several economies delivered neither sustainable industrialisation nor import substitution, notwithstanding structural changes in the composition of the manufacturing sector and the import schedule. Yet, it must be noted that CEPAL diagnosis, and policies associated with it, were neither anti-private capital nor anti-engagement in the international economy, even if they came to be envisaged as such. Perhaps the problem was not cepalismo but the cepalistas or those who self-described as cepalistas. Thirdly, that there is continuing disagreement about when and how industrialisation took root in different parts of Latin America, and how the continent ‘caught-up’ with other regions of the globe.

The articles is organised as follows: the next section locates the ECLA project and decades within a larger stylised chronology; this is followed by a focus on the pre-CEPAL period -as Love would have it, cepalismo before CEPAL; the next substantive section reviews the ECLA project in terms of ideas/ideology, content and delivery; the final section reviews recent re-evaluations of the project and its relevance in the twenty-first century -whither cepalismo and ideas about development in Latin America?

A Stylised Chronology of Industrial Change and Definition of Industry

From contending approaches to long-run economic and social change, a stylised chronology of industrial growth and industrialisation may be elaborated. This simplified chronology acknowledges that the attainment of a modern society founded upon a developed economy has been an enduring objective in Latin America, exercising pensadores and policy-makers intermittently since the revolutions for independence at the beginning of the nineteenth century. The promotion of manufacturing activities was regarded as central to the realisation of independence and ‘progress’ -to employ a much used a contemporary term. Various views as to the most appropriate means of stimulating industrial expansion prevailed: options included direct state aid for manufacturing and a more generalised encouragement of growth that would facilitate individual initiative in the industrial sector alongside investment in other activities. Concern about the subject, and possibly a perception of failure to secure manufacturing on a firm footing, is revealed in ubiquitous reference to fomento, mejoras materiales and industria at mid-nineteenth century, in the proliferation of the term ‘industries’ in official publications of the turn of the nineteenth century, discussion about the respective merits of ‘natural’ and ‘national’ industries at much the same time, in processes of economic internalisation in the second quarter of the twentieth century, and in contending models of import-substitution industrialisation and diversified export-orientated growth in the final third of the twentieth century, with a renewed search for industry-centred growth strategies in the twenty-first century. Changes in contemporary language and in the content and focus of the literature reflect, in turn, shifts in the composition of manufacturing, in the perceived role of industry, in policy-making assumptions, and in schools of historical analyses. All inform the definitions of manufacturing, industrial growth and industrialisation.

Elaborating a generalised framework obviously obscures distinct national or regional differences and divergences, yet it is possible to identify reasonably distinct sub-periods in the industrial history of Latin America. All periods may not be sharply demarcated (nor precisely dated) for each economy, nevertheless, the specific characteristics of each phase and the transitions from one phase to another have a broad continental aspect, certainly amongst the larger and middle ranking economies. Several such periods can be distinguished: (i) the decades immediately following independence, years of sharp re-adjustment for various expressions of colonial manufacturing that also witnessed attempts to establish modern industry, notwithstanding the first ‘de-industrialisation’; (ii) the age of export-led growth from c.l870 to around the First World War associated with institutional modernisation, the development of infrastructure and demand expansion that created a market for consumer and capital goods -part supplied by imports and part by domestic production, in short, export-driven industrial expansion; (iii) the interwar decades, years of increasing volatility in the foreign trade sector and, not least in the 1930s, of increasingly internalised growth that may have signalled autonomous industrialisation in some countries, a process marked by changes in both the scale of manufacturing and the composition of domestic industrial output; (iv) the classic phase of state-directed import-substituting industrialisation dating from the 1940s (or possibly the 1930s) until the 1970s when ‘forced’ industrialisation became a near continental policy goal; (v) the final third of the twentieth century, decades associated with both industrial deepening and de-industrialisation within a context of global re-insertion -a turn towards the market, or a return to the market, as contending ideologies of neo-structuralism and neo-liberalism gave way to the apparently dominant model of economic internationalism and state and macroeconomic re-structuring; (vi) renewed global instability of the early twenty-first century, triggering in some countries a strengthening of the social-market model, and in others an emphasis on radical statism, sometimes entailing a renewed emphasis on ‘national industrialisation’. As indicated above, the sections that follow focus on phases (iii) and (iv).

Autonomous Industrialisation: cepalismo before CEPAL

“Industrialization in Latin America was fact before it was policy, and policy before it was theory”. (Love 1994, 395)

“[…] there was substantial industry in Latin America well before 1930”. (Haber 2006, 537)

Definitions of ‘autonomous’ industrialisation embrace various, often conflicting, assumptions. Some present a general progression from export-driven industrial growth to ‘natural’ industrialisation as the manufacturing sector achieved critical mass and the rhythm of industrial activity was decoupled from the performance of the export sector. Manufacturing became the leading sector, determining rates of growth and impelling structural change in the economy at large. This form of autonomous industrialisation differed from post-1930s import-substituting industrialisation (ISI) in that it was not ‘programmed’ by desarrollista state action.

How widely observed was this autonomous, ‘natural’ industrialisation? It is impossible to construct a continental framework of autonomous industrialisation -and in several cases, it was not observed at all. For some economies, the First World War may have occasioned a shift from industrial growth to industrialisation. Elsewhere, the early decades of the twentieth century saw a process of continued if uneven expansion in manufacturing output with some qualitative developments, principally the rise of impersonal forms of corporate organisation and diversification of production. In yet other economies, the pace of industrial change slackened during the 1910s and 1920s. In Brazil, the War is sometimes presented as accelerating industrial transformation and institutional consolidation in favour of manufacturing, and sometimes depicted as undermining the impetus to endogenous industrialisation facilitated by export-led growth -namely access to easy credit and capital goods imports (Dean 1969; Versiani 1979; Suzigan 1986). It is also confidently asserted that Chile industrialised between 1914 and 1936, namely, that industry assumed the role of lead-sector. But there is disagreement as to whether this process represented a progression to industrialisation per se or was due to crisis in the export sector occasioned by the collapse of nitrate exports at the end of the War (Palma 2000 & 2015). Surprisingly for some, in terms of contribution to Gross Domestic Product (GDP), Colombia may have had the third or fourth largest industrial sector in Latin America by the First World War, though probably even more regionally concentrated than elsewhere (Ospina Vázquez 1955; Palacios 1980). Mexico, to take another example, might have experienced something approaching industrialisation in the last decade or so before the Porfirian state collapsed in 1911 (Beatty 2001; Bortz & Haber 2002). The Argentinian experience is usually presented as one of sustained, if cyclical and sub-sectorally specific, industrial expansion from the 1890s to the 1920s but not of industrialisation; the years 1914-33 have been projected as a period of missed opportunities or ‘delay’ (la gran demora), an era when the potential for ‘natural’ industrialisation existed but was not realised, in part because the state did not consciously work in favour of manufacturing -a view that has been increasingly challenged (or refined) of late (Di Tella & Zymelman 1967; Díaz Alejandro 1970; Barbero & Rocchi 2003; Rocchi 2006; Pineda 2009). Peru's manufacturing sector, on the other hand, witnessed secular decline. Following years of florescent industrial activity between 1891 and 1908, fuelled by an export boom involving a diverse range of commodities, and producer and merchant investment in a broadening spectrum of domestic activities, the next twenty years or so were ones of relapse. Sluggish domestic demand, the negative consequences of exchange stabilisation and reduced protection for local manufactures, plus export volatility, saw a decline in the pace of industrial growth (Thorp & Bertram 1978; Drinot 2000).

Linear projections must be treated with caution. In Argentina, the ratios of agriculture and livestock output to manufactured output had shifted from 2.1:1 in 1900 to 1.3:1 in 1929, confirming that while the index of agricultural production had risen from 29 to 117, the index for manufacturing had increased at a much faster rate from 9 to 46 (1950=100). These trends indicate some restructuring of the economy, notwithstanding the small base from which manufacturing expanded (Díaz Alejandro 1970: 418, 420, 433-4, 449). The late 1920s also witnessed remarkable industrial activity in Mexico, evidencing recovery from the effects of the Revolution as well as representing an advance upon Porfirian achievements. Manufacturing output, which had grown by an average of 3.1 percent per annum during the period 1901-1910 and registered a decline of 0.9 percent a year from 1911 to 1921, expanded at an average annual rate of 3.8 percent between 1922 and 1935, and there was evidence of fairly significant positive institutional change in the manufacturing sector (Solís 1970; Cárdenas 2000; Marichal & Cerutti 1997; Gómez Galvarriato 1998 & 2002). Both in Mexico and Brazil, qualitative changes in manufacturing were reflected in an increased use of electricity (only partly explained by a switch from earlier forms of power such as steam), particularly cheap hydro-electricity. The scale of production also increased. Brazilian industrial output grew erratically, but nonetheless dramatically, between 1914 and 1929. Two interrelated developments in the areas of transport equipment, chemicals and electrical goods point to structural change and more intensive patterns of activity. Influenced by the greater availability of electrical power and domestic demand expansion, the first was an increase in productive capacity, the second, penetration of these sub-sectors by transnational corporate capital (Dean 1969; Suzigan 1986). Four key trends are observable with respect to Chilean industry between 1914 and 1929 and beyond, a period of sustained if uneven expansion. There was a change in the scale of production in favour of larger firms; local manufacturers increased their share of the domestic market by approximately 50 percent; there was a relative decline in the participation of current consumption goods (such as foodstuffs and textiles) in total domestic manufactured output, and an increase in the share of consumer durables, intermediate products and capital goods (like paper, chemicals, machinery and transport equipment); manufacturing increased its relative participation in Gross National Product (GNP) (Palma 2000 & 2015, 320-335). By around 1930, aided by a combination of geography, export-led growth residuals and market responses to opportunities provided by global crises, the modern manufacturing sector in Colombia included the production of household appliances in addition to such established items as textiles (McGreevy 1971; Palacios 1980; Jaramillo-Echeverri, Meisel-Roca & Ramírez-Giraldo 2015).

Between the 1900s and the 1930s, intra-sub-sectoral industrial diversification occurred in the three larger economics and in some middle-sized economies like Chile, Colombia and Uruguay. Such diversification included an expansion of consumer durables output (featuring household electrical goods manufacture and motor vehicle assembly), the production of a broadening mix of intermediate items like chemicals, petroleum and pharmaceuticals, plus a modest deepening of capital goods manufacture. Indeed, in these economies, the consolidation of industries after 1930 is testimony to the scope of a pre-Crash manufacturing base. Yet, it is easy to exaggerate the magnitude of structural change in the early twentieth century, particularly the interwar period. For Haber, quoted at the beginning of this section, despite precocious industrial development since the mid-nineteenth century, the pace of expansion of whatever industries had been founded was very modest (Haber 2006, 539). While some would question this assertion, the growth in manufactured output and installed industrial capacity must be set against the continued dominance of the foreign trade sector in many economies. Agriculture and mining often remained the focus of activity even if industry may be projected as a dynamic -possibly the most dynamic- sector. Against complacent accounts depicting an inexorable progress to an industrial society must be set more cautious interpretations which stress the limits of industrial expansion induced by export-led growth, and the continued technological backwardness of much of the sector, notwithstanding the existence of world-class firms (Thorp 1998, 87-95, 107-120; Bértola & Ocampo 2012, 130-131). The easy phase of export-led industrial expansion based on the processing of rural products, the refining of minerals and the manufacturing of basic consumer goods was drawing to a close in the larger economies. That did not imply an inevitable transition to industrial deepening, whether provoked by external crisis or benefiting from a shift in the stance of the state. Industrialisation required a fundamental restructuring of the social order. Arguably, it was in this respect that the First World War and the global crisis were critical. Exogenous shocks highlighted the dangers of an overdependence upon the external sector. Thus, while it can be affirmed that prior to the 1930s there was no systematic preoccupation with the promotion of industrial development, this does not mean that there were no initiatives with a view to protecting domestic industrial activity and developing certain industries (Versiani 1987; Coatsworth & Williamson 2004; Rocchi 2006).

If the commodity lottery influenced the timing and nature of Latin America's engagement with the global system, it also conditioned domestic linkages and state structure. State structures, and the societal formations within which they were embedded, in turn influenced the capacity of regimes to respond to challenges and opportunities during the interwar decades, not least in the industrial sphere: economic policy hinges on assumptions of developmentalism and sovereignty. If states had actively promoted economic openness and growth during the late nineteenth and early twentieth centuries, did they, at this point, consciously design policies to promote economic change, namely foster manufacturing? By the interwar decades some states were responding to the demands of politicians, entrepreneurs and workers beyond the confines of the traditionally dominant export sector -and were particularly inclined to do so in response to the changing economic environment (Bortz & Haber 2002; Haber 2002; Bértola & Ocampo 2012). Perhaps it is ahistoric to ask whether states might have been proactive. Díaz Alejandro observes that, drawing on the experiences of the 1920s and earlier years, some states were ‘reactive’ (proto-developmental) in the 1930s while others were passive (Díaz Alejandro 2000: 17-49).

Buffeted by external events and domestic forces from the 1920s to the 1940s, the weak, dysfunctional, highly-personalised states typical of parts of Central America, the Caribbean and the interior of South America experienced a rotation of individuals or cliques but were able to ignore the clamour of sporadic, inchoate domestic popular protest, no matter how violent. However, they were unable to construct an active policy response to crises. Perhaps there was no need. Here there were barely economic, let alone political, markets. Elsewhere, though often in the face of acute difficulty, states like the Mexican, Brazilian and, possibly the Chilean, were able to internalise conflict, demonstrating a capacity to frame an autonomous economic programme and, over time, move from reactive to proactive measures. Other states, endowed with well-established institutionalised structures, were able to accommodate regime change within the existing framework while deflecting or muting protest. Pragmatic domestic policies were accommodated within an international economic strategy that changed little. The result, in countries like Argentina and Colombia, was a project that was interventionist but not necessarily consciously developmental.

Simplistic accounts that presented the 1930 depression as provoking regime change everywhere and promoting a new political economy are untenable. Three phases of response to the depression may be identified. Initial policy reactions were fairly consistent from country to country -and consistently minimalist. Over time, objectives diverged: in some countries, policy became more developmental. With regard to state action, the first phase (c.1929-1931/2) was event-driven, disjointed and orthodox. The second phase (c.1932-1934/6) saw greater coherence and some pragmatism in economic policy, though conditioned by lingering assumptions that order might yet be restored to international commodity and financial markets. In the late 1930s, a third phase is identifiable, when some states applied projects that were clearly -and consciously- proto-Keynesian and designed to favour the industrial sector.

Arguably, Argentina best characterised the minimalist and wishful-thinking that influenced state policy during the first phase. Apart from suspending convertibility to stall a haemorrhage of gold in December, 1929, the Yrigoyen government in Buenos Aires adopted a do-nothing approach as the crisis developed around the turn of the year (Rapoport 2012, 143-146, 150-156; Gerchunoff & Llach 2018). Most other administrations (not least those that came to power in 1930) implemented fairly conservative measures. Taxes -mainly import duty surcharges- were raised and attempts made to curb expenditure. There was, too, a credit squeeze: interest rates soared, and loans were called in. (Hence the dictum: “In a recession, all prices fall, except the price of money.”) Consequently, the impact of the crisis was generalised, partly through a contraction in export sector incomes and partly through the state sector which, also experiencing an income crunch, pursued pro-cyclical policies. Although most countries left the Gold Exchange Standard around 1930/31, the measure was presented as a short-term expedient, just as it had been in 1914. And when sterling came off gold in October, 1931, suspension of convertibility hardly appeared radical. The military regime, headed by General Uriburu, which sized power in Argentina in September, 1930, was pledged to return the country to gold (Rapoport 2012, 213-216; Gerchunoff & Llach 2018). After struggling to remain with gold, the Mexican government opted for inconvertibility in July, 1931, around the same time as Chile, but as late as 1933 was seeking to re-institute a silver standard in order to combat monetary anarchy. Moreover, until abandoning gold, all countries behaved in an extremely orthodox fashion: capital flight and loss of gold reserves was accompanied by a sharp contraction in money supply (Díaz Fuentes 1994). And, if many governments were prompt to introduce exchange controls in 1930 and 1931, this was to correct the growing trade imbalance and (if not too late) preserve gold reserves so as to facilitate an orderly return to convertibility at some point in the future. This hardly constituted adventurous proto-Keynesianism. Governments did, however, act to protect the export sector. On assuming office, Vargas in Brazil resurrected coffee defence, abandoned by the ousted administration of Washington Luis in order to balance the budget and avoid inflationary pressure. In Argentina, concordancia governments of the 1930s extended the system of commodity price support that had previously applied only to wheat. Although the mechanisms were quite different, the objective was the same, to preserve the viability of the export sector. The result, of course, was to sustain aggregate demand (or at least prevent further contractions) but this was not the prime motive at the beginning of the decade. If there was a continental -or near continental- response to the onset of the crisis in the early 1930s, it may be characterised as tardy, event-responsive, rooted in the view that the global economy was experiencing a recession rather than a depression. Hence, policy measures were piecemeal, orthodox and largely defensive.

The second phase began around 1932/3. By this stage, the extent of the crisis was beginning to be realised. It was no longer regarded as a temporary disruption to the working of the international commercial and financial order that could be tackled by conventional methods (Love 1994, 406-407). Ad hoc measures that had gradually been applied at the beginning of the decade were now being institutionalised. Foreign exchange was allocated according to a schedule of priorities rather than by availability, on a first-come, first-served basis. ‘Temporary’ tariff hikes and quotas became permanent and, by the middle of the decade, the old multilateral system was being displaced by networks of bilateral trade regimes. These measures now began to benefit mainly manufacturing interests (Coatsworth & Williamson 2004; Haber 2006). Argentina and Brazil were amongst the first countries to implement ‘compensation’ commercial and clearing agreements with Great Britain and Germany respectively. Yet the proliferation of bilateral commercial and clearing agreements was hardly original when Britain had opted for imperial preference in 1932, and in 1933 the US dollar was devalued and the London Conference (the World Monetary Conference) failed to reach agreement to defend the multilateral order. This unorthodox policy shock from the centre galvanised many Latin American administrations to action (Díaz Alejandro 2000, 21). Moreover, while it may not have been obvious to contemporaries, many economies were already on the road to recovery. The depression bottomed-out between 1932 and 1936: aggregate output was starting to grow again and, in some cases, the volume of export production (though not necessarily the actual value of exports) was at or around pre-crisis levels. Perhaps this was a measure of the success of attempts to ‘defend’ the export sector and resultant domestic spin-offs. The guiding principle of this second phase was pragmatic orthodoxy, bounded by the need to be seen to be responsive to powerful domestic sectors, including the industrial lobby, while also accommodating competing overseas commercial and financial interests.

The third phase may be observed by the close of the decade. The timing of the transition can be dated as beginning around 1935-7. Although the Cárdenas sexenio in Mexico began in 1934, the middle years of his presidency -characterised by massive land reform and the expropriation of foreign-owned oil companies- proved to be the most innovative phase (Cárdenas 1987 & 2000). Similarly, the Estado Nôvo in Brazil, explicitly echoing the New Deal in the USA, was launched in 1937 following further political turmoil in 1936, a year when coffee prices again nose-dived. By 1937-8 it was fairly obvious that war was coming in Europe. With the prospect of export price recovery came the possibility of greater freedom in economic decision-making. By the late 1930s it was also clear that international capital markets were not going to re-open and that many countries, not least in Europe, were breaking old rules. The opportunity cost of unorthodoxy declined accordingly. Policy now became more explicitly pro-industry. Following sharp externally induced price falls at the beginning of the decade and given the growing sophistication of exchange control mechanisms, most governments were less anxious about the impact of ‘fiscal delinquency’ and domestic inflation upon the exchange rate. By 1935 Mexico was demonetising silver and embarking on monetary expansionism. Nacional Financiera S.A. (NAFINSA), the state development bank established in 1934, would soon preside over a constellation of sectoral credit agencies designed to foster domestic capital market growth. Nevertheless, only after 1940 would NAFINSA become thoroughly committed to the promotion of manufacturing. The Chilean Corporación de Fomento de la Producción, set up in 1938 initially to promote regional regeneration following an earthquake in the north, soon emerged as a fully-fledged development agency. While there is some dispute in the Brazilianist literature, a persuasive case is made by authors who argue that the Vargas regime only became consciously and systematically pro-industry in 1937, when manufacturing activities were directly targeted. Industrial modernisation became the central objective of the Estado Nôvo (1937-1945), embracing the project to establish an integrated iron and steel complex -Volta Redonda- and state investment in associated areas such as mining and energy generation (Wirth 1970; Villela & Suzigan 1973). National development and industrialisation were central to the ideology and policy of the period -often regarded as one and the same thing (Bresser-Pereira 2009, 62). It may be no coincidence that countries embarking on more explicit pro-manufacturing programmes also sought to bind urban labour to the state. In Mexico and Brazil, a regime of state-controlled trade unions and welfare enhancement (on a modest scale pre-figuring that of the 1946 Peronist administration in Argentina) dovetailed with a macroeconomic strategy in which support for manufacturing became more explicit. Yet it would be incorrect to characterise the overarching policy objectives of the late 1930s as import-substituting industrialisation. On the contrary, ‘economic internalisation’ was the overriding goal. Pro-manufacturing initiatives were subsumed within this larger framework which may be depicted as export-substitution (and export diversification) as much as import-substitution.

‘Economic internalisation’ was pursued almost everywhere. The conjuncture of domestic political and economic pressures -reinforced by external dislocation- accounts for the emphasis on economic internalisation during the Cárdenas sexenio. Closely bound to the USA, the Mexican economy was doubly affected by economic crisis north of the border. As US GDP contracted by around 40 percent, the depression was exported south. In addition, job shrinkage in the USA produced a mass return of migrant workers. A new round of bank failures in the USA in 1933 seemed to indicate that the worst was not yet over. Drastic action was required. Economic radicalism can also be explained by attempts by Cárdenas to construct a political base independent of Calle, the jefe máximo. An emphasis on collective ejidos, mass land expropriation accompanied by compensation payments in government bonds, credit expansion and banking reform, investment in the social and productive infrastructure (education, transports and electrification) and a favourable attitude to wage demands increased aggregate demand and mobilised capital, facilitating the formation of new financial-industrial grupos. The state apparatus was also overhauled, providing for more effective political management and sense of stability (Cárdenas 1987 & 2000; Cerutti 2000). Modest by comparison, the revolución en marcha launched by Liberal president Alfonso López Pumarejo (1934-1948) in Colombia echoed elements of the cardenista programme. As in Mexico, there was to be greater tolerance of organised labour -the rhetoric was of independent unionism- and measures to enhance worker rights. Fiscal and credit reforms were proposed and, in 1936, a new land law which seemed to favour the rights of small farmers and limit latifundismo, was placed on the statute books. In part deriving from these reforms, investment in, and production for, the domestic economy rose markedly (Ocampo 2000 & 2007; Ospina Vásquez 1955 & 1963). Institutional change around this period had a lasting impact. It is probably no coincidence that the Colombian economy demonstrated remarkably stable and near continuous rates of growth for a good half a century or so following the early 1930s. While average rates of growth sustained by Colombia cannot be compared with periodic ‘miracle’ rates manifest elsewhere, Colombian growth was considerably less volatile, facilitating steady progress. This was the result of limited, market-sympathetic, pragmatic, prudent interventionism -or competent macroeconomic management. In terms of business-state relations, economic decision-making was diffuse and regionally and sectorally distributed, thus possibly making for consensus and continuity which favoured business planning about investment and production. In Colombia, industrialisation was business-led and state-supported (Thorp 1991; Thorp 1998; Brando 2012).

There was much ‘learning by doing’ during the 1930s in many countries and, possibly, a recovery of bureaucratic ‘memory’ of action taken during earlier period of external disequilibrium (Thorp 1998, 120-123; Thorp 2000, 2; Bértola & Ocampo 2012, 136-137). Yet, growth during the latter part of the decade often owed much to the recovery of external demand. Indeed, this is suggested in Table 1 (see next section) which shows that, in several cases, the relative size of the export sector in 1938 was not so different from 1928. Nevertheless, the ability of producers to respond to external opportunities was not unaffected by export-defence measures applied earlier. The principal beneficiary of domestic recovery, however, even in Argentina, where government had pursued fairly orthodox monetary and fiscal policies throughout the period, was manufacturing. However, industrial growth was more the result of government policies aimed first to promote economic stability and then a generalised recovery, than measures directly geared to the requirements of manufacturers in several parts of the continent. Only at the very end of the period, and in only a few cases, did manufacturing rise to the top of the policy agenda. In Brazil, the emphasis in policy language certainly bore a more ‘industrial’ gloss after 1937 and influenced both commercial and foreign economic policy. The Alessandri administration in Chile probably pursued more consciously Keynesian and more pro-manufacturing policies than its predecessors. The laissez faire state was distinctly on the wane.

What lessons may be learnt from the history of industrial growth and performance during the period? Was there a phase of ‘autonomous’ industrialisation during the interwar decades, partly occasioned by the disruption of the First World War and global volatility thereafter? The first lesson is that, possibly excepting some Central American republics, substantial industrial capacity had been installed across the continent by the 1930s. The history of modern manufacturing did not begin in 1930, as was sometimes argued. Domestic industry expanded in many of the medium-sized and larger economies during the first globalisation, driven by export-led growth. Producing commodities demanded overseas helped form the domestic market, facilitated capital inflows into a range of manufacturing activities supplying ‘modernising goods’ in response to national demand and, in some cases, modified the policy stance of the state (Bauer 2001, 139-144; Salvucci 2006, 287-291). The second lesson is that governments were becoming more active -the new institutional setting was becoming either more pro-market and growth-enhancing, or even proactive when confronted by macroeconomic volatility and lobbying by special interest groups that included industrial entrepreneurs (Dye 2006; Haber 2006). The third lesson is that the new arrangement was fairly successful in promoting (certainly presiding over) economic recovery and structural change, especially during the interwar period, notwithstanding institutionalists´ preoccupatons about state distortion of market signals and potential for rent-seeking. Perhaps this was because policy remained fairly pragmatic. As yet, there was no ideological commitment to state-led ‘forced’ industrialisation, but the significance of the role of manufacturing in promoting recovery in the interwar decades and potential to foster national development was being recognised. Learning by doing may have meant learning from mistakes as well as successes. Nevertheless, arguments of growing state competence and bureaucratic outreach can be exaggerated. As in the case of Brazil, though there may have been the will, the means weren’t necessarily there to implement development projects (Draibe 1985, 155-156).

State-directed ‘Forced’ Industrialisation

Two main strategic conclusions of this [centre-periphery] analysis followed. First, it was obvious that the modernization and development of these countries required industrialization, as the industrial sector was the bearer of technological progress, and as the eventual exportation of manufactures would redress the disadvantage experienced by primary product exporters in international trade. Second, agrarian reform and rural modernization were also essential in order to overcome the historical backwardness and exploitation prevalent in the agricultural hinterlands. The main public policy conclusion was that, as in the contemporary experience of the developed and socialist countries, the state had to play a central role in bringing about both processes: industrialization and rural transformation and modernization. (Sunkel 1993, 20-21)

“This term (state-lead industrialisation) combines the two main characteristics of this process: an increasing focus on industrialization as a mainstay of development and a considerable expansion of the scope of state action in economic and social affairs.” (Bértola & Ocampo 2012, 138)

“[…] while defenders of ISI saw the policy as working to develop industries with higher value added over time (and with higher productivity rates), its critics saw flaws in the implementation of such stages.” (della Paolera, Durran & Musacchio 2015, 4)

Arguably, it was the onset of the Second War that gave a narrower pro-industry and import-substituting industrialisation emphasis to policy, possibly signalling the success of industrial expansion in the 1930s. The ‘golden era’ of industrialization was about to dawn for many economies, marked by impressive rates of growth in manufacturing output and catch-up with the industrialised countries of the Europe and Asia and the USA (della Paolera, Durán & Musacchio 2015, 14-15 [Tables 1 & 2]). If industrialisation was a function of sectional interest policy-pleading, there was now a larger pro-manufacturing lobby. As indicated above, during the latter part of the Estado Nôvo in Brazil, support for heavy industry had become even more explicit. The Avila Camacho government in Mexico, which assumed office in 1940, was both pro-business and pro-industry. And, in 1943, the military clique that overturned the discredited concordancia regime in Argentina was, exercised by events in Brazil, determined to promote strategic, heavy industrialisation. Moreover, by the end of the 1940s, ISI was dignified by ideology. As suggested in the opening quotations, CEPAL provided an intellectual justification for a co-ordinated programme of forced, or state-directed, industrialisation. And, there was a larger base on which to build, as illustrated by Table 2 which shows rapid industrial growth (if not industrialisation). Mexico and Colombia present the highest average rates of growth of industrial output between 1932 and 1939, though this was from a relatively small base, particularly in the case of Colombia. Manufacturing accounted for just over 6 percent of GDP in 1930 in Colombia and 14 percent in Mexico. Although the rate of industrial growth in Brazil was somewhat above the Argentinian figure, this must be set against the relative sizes of the manufacturing sector in the two countries. In 1930, the share of manufacturing in Argentinian GDP was almost twice the Brazilian. Yet, despite high output growth of over 7 percent per annum, the contribution of manufacturing to Argentinian GDP was virtually the same in 1940 as in 1930. Similarly, in the Brazilian and Chilean cases, despite annual average rates of industrial output growth above 7 percent, the share of manufacturing increased by only a few percentage points between 1930 and 1940 -just over three percent for Brazil and not quite four percent for Chile. Again, this suggests that policy was more effective at internalising growth mechanisms than fostering structural change, namely industrialisation. Indeed, for several countries, the ratio of export earning to GDP was much the same in 1928 and 1938. The exceptions are Argentina and Mexico, a divergence which may in part be explained by the impact of a sharp fall in wheat prices on total Argentinian exports and oil sector dislocation in Mexico.

Table 1 Share of Manufacturing in GDP (%) 

Argentina Brazil Chile Colombia Mexico Latin America USA
1920 17.4 10 40.2
1930 22.8 11.7 7.9 6.2 14.2 8.9 42.1
1940 22.7 15.0 11.8 9.1 16.6 12.1 42.2
1950 23.7 21.2 23.1 13.5 18.3 18.7 39.8
1960 26.5 26.3 24.8 16.7 19.5 21.3 38.7
1970 28.8 28.4 27.2 17.5 22.8 25.1 38.3
1980 25.3 30.2 24.2 18.3 19.1 25.4 35.6
1990 21.6 27.9 21.7 22.1 22.8 23.4 33.7
2000 20.1 25.2 17.1 19.6 19.7 22.3 30.1

Source: Statistical Abstract of Latin America elaborated from ECLA (AC) data.

In contrast, between 1940 and 1950, the contribution of manufacturing to GDP rose remarkably in Brazil, Chile and Colombia. Respectively around a half and a third of the relative size of Argentinian industry in 1930, Brazilian and Chilean manufacturing contributed almost the same share of national GDP as Argentinian manufacturing by 1950. Between 1930 and 1950 the contribution of manufacturing to Colombia’s GDP more than doubled but was still only two-thirds of the Argentinian figure. What construction can be placed on structural change of this order?

Cepalista analyses and prescriptions fell on fertile ground after 1948 when the Commission was established (Bielschowsky 2016, 7-44). Negative views about the terms of trade encountered by commodity-exporting economies seemed to be validated by the recent historical experience of Latin America. The Second World War also confirmed structural changes in the global economy which exacerbated the problems of primary producers. The world economy was now centred on the USA, a mature economy with a huge productivity advantage and a rising propensity to export coupled with limited import requirements reinforced by strong protectionist tendencies. The congruence of experience and theory was a winning combination that contributed to the rapid diffusion of ECLA developmentalism amongst policy-making elites in Latin America. If assessments of the external environment were negative, there were grounds for domestic optimism. Learning-by-doing during the interwar period and the Second World War meant that several administrations were prepared to embrace an even more interventionist approach. CEPAL provided both the justification and the design to do so. Post- depression recovery, growth in manufactured output and economic and political institution-building were interpreted as signalling state competence.

The main policy instruments associated with ECLA developmentalism were exchange control (often manifest in multiple exchange rates that gave preference to the manufacturing sector), protectionism (non-tariff barriers to trade and exchange regulations were employed in conjunction with, sometime in preference to, discriminatory duties) and forced savings. Overvalued, but not necessarily stable, exchange rates prevailed for much of the period and were consistently applied to the advantage of the industrial sector. While only Mexico managed to defend a stable exchange rate throughout the classic period of ISI, the repeated devaluations that occurred elsewhere hardly benefited commodity producers as devaluations were accompanied by windfall taxes on exporters. This was consistent with cepalista trade theory which argued that markets for exports were not price responsive. Devaluation, followed by windfall taxes on exports, was also consistent with the regime of exchange and export profit ‘nationalisation’ and the distortion of the domestic terms of trade in favour of the urban industrial sector. As the principal source of foreign exchange, the export sector was consistently squeezed by state agencies. Inflation was the main, but not the exclusive, mechanism of forced savings. Having looted the social insurance funds (caixas) to finance the construction of the Volta Redonda iron and steel complex, regimes in Brazil and elsewhere learnt to milk -politically as well as economically- the social security system (Mesa-Lago 2008). As a growing proportion of the urban white-and blue-collar workforce was brought within the scope of the social insurance net and while the funds remained in surplus, they were an important source of forced saving (Mesa-Lago 1991, 186). The most sophisticated system of forced savings was devised in Brazil during the miracle years, a period of relatively low inflation when all formal sector workers were compelled to contribute to social insurance funds, the national housing bank (its resources were used to finance road building in Amazonia) and indexed individual savings accounts. Coupled with social repression and income concentration, mechanisms like inflation and social insurance ensured that by far the greater part of investment was financed by domestic savings in countries like Brazil (Suzigan & Villela 1997).

In addition to underlying assumptions of bureaucratic competence implicit in the socio-economic measures identified above, cepalista policy recommendations were also predicated on a belief in the existence of an heroic national entrepreneuriat. Cepalismo may have been interventionist and statist, but it was not anti-business. The role of the state was to insulate and nurture domestic entrepreneurial talent. The state was to serve as an intermediary between new businesses and an unfavourable environment, sheltering firms from unfair competition and providing access to essential inputs, not least capital and technology, and serve as a conduit for aid from international agencies (Schneider 2004, 31-36). There was also the presumption that some countries might emerge as exporters of basic wage goods. Drawing on the W.A. Lewis thesis of a modern, capitalist sector developing on the basis of unlimited supplies of labour siphoned from the ‘traditional’ and on the evidence of installed manufacturing capacity and intra-regional trade in manufactures during the war, the development of industrial exports seemed to be on the agenda. Theoretically coherent, these expectations acknowledged that re-structuring the Latin American economies would remain import-dependent in the medium-term. Although they only assumed concrete form subsequently, the market-orientation of ECLA developmentalism was also confirmed by projects such as regional integration and agrarian reform (Thorp 1998, 149-157). Regional integration was rooted in concepts of efficiency and ‘fair’ competition. Economic integration would facilitate the emergence of large-scale, efficient firms exposed to the rigours of competition from producers in neighbouring countries but still protected from unequal competition in the regional marketplace by overseas conglomerates. The emphasis on agrarian reform also reflected, among other concerns, recognition that growth and efficiency were market-size constrained, though in this case the emphasis was on qualitative deepening rather than quantitative expansion.

What was the outcome? First, state planning often became more effective, certainly more informed as the result of systematic data collection. The Targets Plan (Plano de Metas) of the Kubitschek presidency (1956-1961) is generally reckoned to have been the first effective experience of industrial development planning in Brazil, deriving from the strong policy consensus in favour of modernisation and industrialisation (Bielschowsky 1998; Rossoto Ioris & Rossoto Ioris 2013). It was one of the earliest integrated development strategies focused on industrialisation to be implemented in Latin America (Evans 1979, 55-60; Kaufman 1990, 125-126). In Mexico, with new fiscal and exchange policies in the early 1950s and the inauguration of ‘stabilising development’, state planning became more flexible, reflecting closer relations between the state and business sectors (Cerutti 2000, 161; Schneider 2004, 69-72). Indeed, planning became both subtler and more sophisticated at sectoral and macro level (Hamilton 1982; Cárdenas 1994). It appeared that Mexico had achieved a Rostovian ‘take-off’ (FitzGerald 2000, 213). In Argentina, the establishment of the National Development Council (CONADE) may also be interpreted as an attempt to ‘plan’ (or programme) the economy out of stop-go cycles though sectorally co-ordinated industrialisation (Díaz Alejandro 1970; P. Lewis 1990, 282, 292, 317). Perhaps the high-point of co-ordinated, planned industrialisation was the radical corporate-syndicalist experiment essayed in Peru after the 1968 golpe (FitzGerald 1976; Thorp & Bertram 1978). Second, there was absolute job growth (notwithstanding an emphasis on capital-intensive methods of production) and diversification in the structure of manufactured output (ECLA 1967). And, as indicated in Table 1, the participation of manufacturing in GDP rose. This suggests both welfare and productivity gains. Labour productivity and total factor productivity (TFP) increased. Output per worker rose fairly consistently from the 1930s to the 1970s, with TFP rising steeply between the late 1930s and the late 1970s -partly closing the productivity gap with the advanced economies (Hofman 2000, 34-35, 112-119; Astorga, Bergés & FitzGerald 2011, 204-207; Bértola & Ocampo 2012, 174-176). Although real wages lagged those of the North Atlantic economies, for workers there were important welfare gains: job security, wages and working conditions in the modern manufacturing sector, by this stage dominated by transnational corporations (TNCs), were better than in traditional industries and much better than in agriculture. There was a massive reduction in absolute levels of poverty between the 1940s and the 1970s (Astorga, Bergés & FitzGerald 2005; Prados de la Escosura 2007). These gains explain why consumer durables assumed the predominant share of manufactured output in all the large- and medium-sized economies. In virtually all these republics, rates of growth in value added in the manufacturing sector were highest during the classic period of ISI (1945-1972) than for the immediately preceding and subsequent periods (Thorp 1998, 322; della Paolera, Durán Amorocho & Musacchio 2015, 13).

Further evidence of structural change is also suggested by an increase in manufactured exports, though the evidence may be challenged. The value of industrial exports from Brazil rose from a figure that, in 1955, hardly registered in the export schedule to reach 14 percent of total exports by 1970. Over the same period, the value of manufactures exported from Mexico increased from 12 percent to 33 percent of total trade -a proportion that was not so far from that achieved by some of the East Asian newly-industrialising countries (NICs) a few years earlier (Gereffi & Wyman 1990, 15). Positive constructions placed upon this achievement must be qualified by reference to the share of exports in manufactured output and the nature of the trade. Although manufactured exports may have accounted for a growing proportion of total exports, manufactured exports as a share of total industrial output remained low. In the Argentinian case, less than one percent of total manufactured output was exported in 1960: the figure for 1973 was 3,6 percent. Between 1960 and 1973 the share of Chilean manufactured output exported actually fell, from 3.0 percent to 2.5 percent. The figures for Brazil for the two years were 0.4 percent and 4.4 percent respectively; for Mexico 2.6 percent and 4.4 percent; for Colombia, in this respect the best performing economy, 0.7 and 7.5 percent (Kaufman 1990, 130). This ‘inward-bias’ contrasts with data for export-led industrial deepening in East Asia. In the 1970s it became fashionable to cast further doubt on the dynamics (and the dynamic affects) of even this modest export performance. Rather than marking increased efficiency and international competitiveness on the part of industrial enterprises in Latin America, the growth in manufactured exports simply reflected the ‘transnationalisation’ of world trade in manufactures which was becoming intra-corporate rather than inter-country (Jenkins 1987, 95-122, 165-194; 1991). However, it must be remembered that ‘Latin American’ TNCs were participating in this ‘transnationalisation’ of global trade: not all corporations were ‘foreign’ (Díaz Alejandro 1977; Jenkins 1987, 144-164; Katz & Kosacoff 1983; Bisang, Fuchs, & Kosacoff 1992). While external operations by Latin American TNCs hardly compare with that of overseas corporations in Latin America, foreign activities by these enterprises supports arguments both about the transnationalisation of business and the entrepreneurial behaviour of manufacturing firms founded by Latin American grupos. In addition, it must be remembered that markets in Latin America were larger and, possibly deeper, than in the newly industrialising economies of East Asia. In East Asia, the ‘easy’ phase of early industrialisation focused on domestic markets was short -often barely an option, even if it may be argued that a focus on internal markets lasted ‘too long’ in the larger economies of Latin America (Jenkins 1991; Grinberg 2011).

From a ‘neo-liberal’ perspective, the results of classic cepalismo are easily disparaged. Yet, it is worth repeating that the achievements were substantial. By its own history, economic performance and industrial growth during the ISI period was impressive in Latin America and compares not unfavourably with that of the advanced economies during similar phases of their development (Brando 2012, 33-35; della Paolera, Durán Amorocho & Musacchio 2015, 2).

Table 2 Average Annual Growth Rates of Industrial Added Value 

1950-74 1974-80 1980-90 1990-94
Argentina 4.9 -0.6 -1.4 6.9
Brazil 8.7 6.7 -0.2 2.8
Chile 4.4 1.2 2.6 6.3
Colombia 6.7 4.0 2.9 3.9
Mexico 7.4 6.2 2.0 2.3
Peru 7.0 1.8 -1.9 5.6
Uruguay 2.4 4.9 -1.0 -1.3
Venezuela 7.8 5.0 1.9 1.8
Average 6.2 3.7 0.5 3.5

Source: Elaborated from Benavente, Crespi, Katz & Stumpo 1996, 57.

Table 2 captures the qualitative and structural changes in manufacturing referred to above. The increase in industrial valued added was noteworthy in the heyday of the CEPAL model between 1950 and 1974, and manufacturing growth drove the productive sectors. Apart from Argentina and Chile, industrial growth was also substantial during the latter part of the 1970s. Most economies realised historically high rates of economic growth during the period. Manufacturing output grew absolutely and relatively: there was industrial deepening and productivity gains associated with more intensive patterns of production and technology absorption. There were, too, major organisational changes. Paralleling the proliferation of state agencies and enhanced state competence, and possibly consistent with the rent-seeking construction placed on forced industrialisation by some of its critics, there was a qualitative and quantitative growth of business organisations (Ffrench-Davis 1973; Draibe 1985; P. Lewis 1990; Sikkink 1991; Cárdenas 1994; Leopoldi 1994; Schneider 2004). The new organisational setting points to the institutionalisation of state-business relations -the emergence of the so-called triple alliance of state, domestic and transnational corporate capitalism (Evans 1979; Hewlett & Weinert 1982; Gereffi & Wyman 1990). Rarely stable, the new institutional setting also witnessed substantial absolute welfare gains (Urrutia 1991; Maddison 1991; Albala-Bertrand 1993; Astorga, Bergés & FitzGerald 2005). The presence of the middle classes increased, urban industrial labour became more organised (usually closely supervised by the state) and, compatible with industrialisation, the urban economy grew exponentially. There was quantitative and qualitative market growth despite the increasing informality and rising inequality; as already stated, there was a massive reduction in absolute poverty, even while relative inequality grew (Astorga, Bergés & FitzGerald 2005; Prados de la Escosura 2015).

Nevertheless, as the process of import-substitution began to encounter problems in the late 1950s and early 1960s, criticisms of ECLA policy prescriptions and the analysis on which they were based multiplied. Dependistas observed that import-substituting strategies had resulted in distorted, dependent industrial growth which had deepened Latin American underdevelopment and induced a new form of dependence. Latin American manufacturing -demonstrably the most profitable sector of the economy- was unbalanced and externally rather than domestically integrated. Production was capital-intensive and skewed towards the manufacture of consumer durables -motor-mechanical, electrical and pharmaceutical goods. This necessitated the perpetuation of inequitable patterns of income distribution. Above all, the sector was dominated by an oligopoly of TNCs that, importing technology and components, financed operations from locally accumulated sources and siphoned profits overseas. Nationalists, too, were antagonised by the import-dependence and low endogenous multiplier associated with foreign dominance of the industrial sector. Like dependistas, they lamented the inculcation of inappropriate patterns of consumption. They were also antagonised by crowding out of local businesses and a tendency, noted in the latter part of the 1960s, for foreign conglomerates to escape from the consumer durables ghetto, to which they had been confined for much of the post-1930s decades, to penetrate the production of wage goods (tobacco products, textiles and domestically consumed foodstuffs), hitherto largely the preserve of locally-owned firms. Liberals (and later neo-liberals) observed rent-seeking, a product of over-zealous regulation, and macroeconomic instability triggered by demand creation -monetary expansion and easy credit, notoriously reflected in inflation and balance of payments crises. Liberals also pointed to the misplaced pessimism of cepalismo: world trade had grown rapidly after the 1940s and international liquidity increased after the 1950s. Yet, for liberals, the key criticism was competitive failure. Although by the 1960s the export sector was no longer the prime generator of savings in most Latin American economies, exports remained the principal earner of foreign exchange: domestic manufacture had failed to service its import needs.

Some of these criticisms own much to ideology and hindsight and little to constructive reflection based on an appreciation of original, ‘authentic’ cepalista strategy. Problems of outcome may be due more to selective application rather than theory per se: the difficulty lay with the cepalistas not cepalismo (Sunkel 1993). Díaz Alejandro shows that, in several branches of manufacturing in Argentina, there was little growth in the share of apparent consumption captured by domestic producers between the 1920s and 1950. Already by the interwar period, local firms were supplying the greater share of the market in the production of basic commodities such as ceramics, clothing, publishing and printing, paper, and tobacco. Indeed, by the onset of the Depression, domestic firms already held almost half the market for manufactures. In short, the ‘easy’ phase of ISI was already exhausted before the onset of (state-led) ISI. The way forward lay in vertical industrial integration or production for export. What occurred, however, was horizontal diversification -the production of more of the same for a highly-protected home market (Díaz Alejandro 1970, 220-254). At mid-century, import-substitution beginning with the manufacture of wage goods was only viable for any length of time in economies, as illustrated in Table 1, where the contribution of manufacturing remained low until the 1940s. Elsewhere, export-led industrial growth or autonomous industrial expansion had already eroded this option.

These gains (and costs) underscore the importance of institutions and getting institutions ‘right’: industrialisation of the order of magnitude observed during the period of forced, inward-looking development is inconceivable without state action. While, in the 1980s and 1990s, it became fashionable to criticise the ‘dead-end state-led closed-economy model’ essayed in Latin America with the vibrant ‘market-friendly open-economy model’ favoured in parts of East Asia, many such comparisons may be viewed as ahistoric or myopic -as has been recognised by more recent evaluations (Naya et al. 1989; Gereffi & Wyman 1990; Jenkins 1991; Chang 1994; Mesquita Moreira 1995; Grinberg 2011). The domestic and global environments confronting the soon-to-be applauded Newly Industrialized Countries (NICs) of East Asia in the 1950s, 1960s and 1970 were quite different from those encountered by Latin America. The literature that devoted considerable attention to the performance and structure of the industrial sector in East Asia and drew adverse contrasts with Latin America in the post-Second World War period misses two critical points: the scale of external assistance (soft aid and commercial preference) available to some insular and peninsular Asian economies during the Cold War, and the institutional shock associated with defeat and occupation (or liberation) in the Pacific War. This does not mean that Latin American countries would not have benefited from thorough-going agrarian reform, substantial social and economic investment (particularly in education and transport), flexibility in policy application (especially tariff protection which could have been more selective and contained phase-out provisions) and greater state efficiency (not least fiscal reform and cohesion within the state sector). Indeed, many such strategies had featured prominently in early (‘authentic’) cepalista policy recommendations (Thorp 1998, 149-157; Bértola & Ocampo 2012, 202).

As indicated above, over time there was an accretion of meta-strategies to the CEPAL development project. By the mid-1950s, regional economic integration and agrarian reform were explicitly on the agenda. Industrialisation, rural modernisation and regional integration came to be regarded in some policy-making circles as complementary and mutually re-enforcing. The question was whether or not such accretions as agrarian reform and economic integration could be delivered. There were two models of regional integration which can conveniently be described as the ‘free-trade’ approach and the ‘common-market’ approach. The former regarded increasing trade (bounded competition among relatively equal trading partners) as the route to integration. The latter regarded structural harmonisation as a pre-requisite for closer commercial relations. The Latin American Free Trade Area (LAFTA), on the one hand, and the Central American Common Market (CACM) and the Andean Pact (AP), on the other, epitomised these approaches. Administratively, the free-trade approach was assumed to be ‘easier’, the structural harmonisation approach as politically more difficult. Perhaps because the Central American economies were smaller and states more amenable, not least to the influence of CEPAL, and because some countries were already implementing trade agreements, the Common Market Approach seemed realisable. In South America (and Mexico), there was no question of ‘prior’ harmonisation; the ‘gradualist’ free-trade mechanism was the only politically feasible option (Bulmer-Thomas 1995, 292-298; Bértola & Ocampo 2012, 160-161). LAFTA and the CACM were both launched in 1960; in the case of Central America, pre-existing bilateral commercial agreements between pairs served as a platform for harmonisation. Progress in both cases was initially fairly rapid, but soon hit the buffers due to political disagreement. Partly in response to frustration with the LAFTA model, and initially learning from the Common Market approach, the Andean Pact was formed in 1969.3 Yet disagreement about the pace of harmonisation, including the role of the state in the economy and regimes governing direct foreign investment in the industrial sector, soon disrupted AP integration. By the early 1980s, the debt/loan crisis and associated political turmoil ended these early attempts at regional economic integration. Yet it is instructive that when integrationist projects were revived later in the decade, and progressed during the 1990s, the Common Market approach was favoured.

Projects of Agrarian Reform fared even worse. Policymakers recognised that the backwardness of rural sector producing for the home market was a major cause of inflation and sluggish demand growth; there was insufficient political will to drive through the changes required. Landowners were too powerful; states were weak; industrialists objected to the reallocation of state investment funds to the rural sector that would have been required to raise productivity and efficiency in agriculture. At best, political palliatives like modest land reform were assayed; there was little appetite for thoroughgoing agrarian reform (Bértola & Ocampo 2012, 1960-1967).4 Before the 1980s, there was little sustained improvement in agricultural productivity outside the export commodity sector. The increases in agricultural production and agrarian productivity for the home market that were observed during the period tended to be associated with the opening up of new frontier regions where such existed (particularly state land), and the transnationalisation and corporatisation of large estates. This model of rural modernisation was some distance from the social and economic transformation of the countryside envisaged by the promoters of agrarian reform which viewed the countryside as both an efficient provisioner of urban markets for food and industrial inputs and as a growing market for capital and consumer goods produced by national industry.

In most cases during the classic period of ISI, states were fiscally weak and politically weak, including many cases of authoritarian rule. If there were exceptions, it was probably in the cases of Brazil and Mexico (Hewlett & Weinert 1982). The Mexican state was relatively institutionally robust -certainly durable from the 1930s to the 1980s- especially during the period of ‘stabilising development’. This was a state structure built on the political kudos delivered by land reform and some of the economic gains associated with agrarian reform and the Green Revolution. Such institutional robustness was critical to sustaining a commitment to industrialisation which combined state dominance in some sectors with state-direction and ‘space’ for the market in other branches of manufacturing. And all this combined with a remarkable degree of macroeconomic stability from the 1950s to the early 1980s. Rural productivity gains triggered by agrarian modernisation facilitated a growth of ‘high-value’ commodity exports which compensated for the decline in oil exports in particular, and covered import requirements, debt service and the remittances of foreign corporations. In Brazil, during a similar period, the central state accumulated influence and authority, at least sufficient to maintain consensus about a commitment to industrialisation. There was disagreement among governing factions about the means to the end -sponsoring industrial expansion within an ‘open’ economy, or ‘closing’ the economy- but not about the overriding objective of industrialisation per se. And there was a growing realisation that delivering ‘development’ based on industrialisation required rural modernisation and productivity advances in the manufacturing sector.

Yet many advances achieved during the cepalista decades are only now being fully explored -possibly, even fully appreciated, as are some of the ‘false’ contrasts or comparisons made with East Asia. In part, this may be because positive trajectories of the late cepalista, sometimes associated with refinements of the model, were cut short by the lurch to the right in the 1970s, the abrupt opening of several economies associated first with bureaucratic authoritarian regimes and later the consolidation of neo-liberal model, followed by the catastrophe of the debt/loan crisis of the 1980s. As already indicated, there was reasonable evidence of positive structural changes in the manufacturing sector by the late 1960s or early 1970s. These involved a rising participation of domestically produced capital and intermediate goods in total output, and a growing participation of manufactures in exports, trends under-pinned by a modest closure in the productivity gap with some of the more advanced economies -even if not the USA, hardly a favourable comparator for virtually any economy during the period. This within the context of near linear growth in industrial output in most of the large and medium Latin American economies between the 1940s or the 1950s and the mid-1970s -Chile was probably the principal exception with a sustained, massive surge in output after the mid-1970s that eclipsed trends during the classic ECLA period (Benavente et al. 1996; MOxLAD). Moreover, challenging the coming analysis of the time associated, O’Donnell, Serra and others argue persuasively that an authoritarian, repressive socio-political context was not a prerequisite for industrial deepening a la East Asia (O´Donnell 1973 & 1991; Serra 1979; Hirschman 1981; Kaufmann 1985). In so doing, such work contributes a corrective to negative near-contemporary assessments of what had been achieved by the 1970s -and what might have been delivered had the modified cepalista strategy continued. It also demonstrates that the ‘open politics’ of many industrialising Latin American economies during the period was less crabbed and repressive than in newly industrialising East Asian economies. The much-vaunted East Asian model had a dark side that tended to be down-played by many drawing negative comparisons with Latin America (Grinberg 2011). At the time, the democratic deficit was greater in the Asian NICs than the Latin American.

This is not to deny that the ISI model was running out of steam by the 1970s, marked by the macroeconomic instability employed by bureaucratic authoritarian regimes to justify a closing down of politics as a prelude to economic reconstruction. The costs of inward-development, as it occurred, were indeed considerable (Taylor 1998). Yet it must also be remembered that the evolving strategy of development advanced by CEPAL consisted of three interconnected, mutually supporting strands: state-directed industrialisation, agrarian reform and regional economic integration. If industrialisation was intended to restructure economies in order to deliver sustainable development, social and economic dimensions of agrarian reform were envisaged as resolving bottlenecks associated with inequality and low productivity, while regional integration would reduce scale constrains imposed by relatively small domestic markets. By focussing narrowly on industrialisation, to the neglect of other components of the model, as Sunkel implies, cepalistas failed to deliver authentic cepalismo (1993). How could the model work if only one strand was fully articulated -accepting that political opposition to agrarian reform and regional integration was considerable. Had the lurch to the right and ensuing financial crisis not occurred, what outcomes might have resulted had authentic cepalismo been delivered remains among the most intriguing imponderables of the recent economic and political history of Latin America.

In at least one respect, however, the criticism of ‘Latin America’ through the ‘East Asia’ optic is telling. As cepalistas and others recognised, rapid (or forced) industrialisation required the accumulation of rents. Rents are good for growth. In cases of late industrialisation catch-up, rents may be critical (Amsden 2001, 1-13; Chang 2003, 19-51). Accepting that such may not be the only route to development or, indeed, everywhere realisable, first-best institutional arrangements like those vaunted notoriously by Acemoglu, Johnson & Robinson (2001) might be preferred. Yet, as North, Summerhill and Weingast (2000) recognised, second-best institution may have to suffice. In circumstances where the state is weak or predatory -or both, cronyism can kick-start industrial growth. Crony capitalism generates rents and enough credible commitment to trigger investment in domestic industrial activity in fragile polities by would-be entrepreneurs. Cronyism may explain the paradox of growth amid political instability, not least during the core periods of industrial growth and industrialisation considered in this article (Haber, Razo & Maurer 2003, 18-20). Small groups of asset holders -in this instance industrial organisations- are provided with special rights and entitlements. These entitlements facilitate the extraction of rents and encourage investment in manufacturing (Haber 2002, xviii). Irrespective of whether rents are good or bad for growth, in such cases, rents are good for catch-up industrial growth and industrialisation. Of course, making the transition from crony capitalism to competitive capitalism is difficult, as the experience of several economies in Latin America in the 1960s demonstrates. Through such devices as inflation, forced savings, protection (tariffs and non-tariff barriers) and differential pricing, states in many parts of Latin America proved adept at accumulating rents and effecting transfers to the industrial sector; they were less efficient than many states in East Asia (and some in South-East Asia) in allocating rents efficiently (Amsden 2001, 1-28; Chang 2003, 132-135; Grinberg 2011, 9-35). In East Asia, there was monitored rent transfer and discipline, with implications for productivity and competitivity; in Latin America, transfer occurred with little discipline or monitoring. The question is no longer whether there was significant industrialisation across the major Latin American economies during the CEPAL decades. As show by Bénétrix, O’Rourke & Williamson (2012, 14-15), Brando (2012, 33-35) and della Paolera, Durán Amorocho & Musacchio (2015, 5-14), there was both substantial industrialisation and industrial catch-up. Indeed, even Williamson acknowledges that there was some (admittedly delayed) catch-up (2011, 199-214). The question now is whether there could have been ‘better’ industrialisation, or better industrialisation ‘earlier’ in the case of some of these economies.

Conclusion

According to any definition, by the beginning of the twenty-first century, Latin America is largely a continent of industrialised, urban economies. Modern units dominate the manufacturing sector and many industrial firms are exposed to the rigours of international competition. Could the post-1970s industrial expansion of countries like Brazil and Mexico have been possible without the manufacturing platform established during the classic age of import-substituting industrialisation between c.1945 and 1972/3? Could productivity gains have been accomplished without sympathetic state action? It seems unlikely. But when did the major rupture occur? With the advantage of new empirical research and the further development of theory, it is clear that a major break occurred in the 1880s, not in 1929/30, nor at Independence. Before the 1880s, in most countries, appropriate institutions were simply not in place. Markets were shallow and factors scarce. Moreover, markets may have been an essential requirement for industrial growth, but their existence did not guarantee an inevitable progress towards industrialisation. That required both capital availability, entrepreneurial commitment and an environment that made it rational to invest long rather than favour liquidity. Yet, if the evidence in favour of export-led industrial growth in the late nineteenth and early twentieth centuries is compelling, this is not to say that such a course was the only possibility, nor that manufacturing would have continued to expand after the 1940s, during the long postwar boom, had the Latin American economies remained ‘open’.

An interesting feature of the literature on manufacturing in Latin America is the way that the geographical and chronological focus -as well as the content- of the debate has shifted. The discussion about proto-industrialisation (and the de-industrialisation of free trade) is largely confined to Mexico and some Andean regions. Southern South America and, to a lesser extent, Mexico figure prominently in accounts of export-led industrial growth. While inward-directed, import-substituting industrialisation may have occurred virtually everywhere, certainly among the large- and medium-size economies, the debate about industrial policy is particularly rich for Brazil and Mexico. This indicates that the controversy about the ‘what’ (rather than the ‘how’) continues: was it states or markets that made manufacturing? What was the cost of protectionism and interventionism? What would have been the price of limited state action?

Evidence points to the rapid growth of industry during period of openness around the turn of the nineteenth century and towards the end of the twentieth. This does not mean that industrialisation was everywhere possible without state action and begs the issue of whether industrialisation equates with development. The key point, however, is that there was a groundswell in favour of programmed, government-directed industrialisation in many countries by the interwar period. The pro-industrial alliance embraced many segments of society. Moreover, the global economy was not always dynamic or buoyant. World trade may have grown rapidly in value and volume from the 1840s to the 1900s and from the 1940s to the 1970s, but this was not the case at the beginning of the nineteenth century or during the interwar period. In addition, there was increasing systemic volatility towards the end of both periods of rapid growth in world trade. Institutionalists, old and new, would also accept that there is more than one route to industrialisation, that conditions changed after the appearance of the first industrialised economy, and that agencies -banks, corporations and states- have an important role to play in late industrialising economies such as the Latin American. The time is indeed ripe for revisiting the industrialisation of Latin America in the twentieth century, and for a re-evaluation of what was achieved during the CEPAL decades.

References

Acemoglu, Daron, Simon Johnson & James A. Robinson. 2001. “The Colonial Origins of Comparative Development: An Empirical Investigation.” American Economic Review 91(5): 1369-1401. [ Links ]

Albala-Bertrand, José Miguel. 1993. “Evolution of Aggregate Welfare and Development Indicators in Latin America and OECD, 1950-85.” In Welfare, Poverty and Development in Latin America, edited by Christopher Abel & Colin M. Lewis, 33-47. Oxford: St Antony’s/Macmillan. [ Links ]

Amsden, Alice H. 1989. Asia’s Next Giant: South Korea and Late Industrialization. New York: Oxford University Press. [ Links ]

Amsden, Alice H. 2001. The Rise of the ‘Rest’: Challenges to the West from Late-Industrializing Economies. New York: Oxford University Press. [ Links ]

Astorga, Pablo, Ame Bergés & E.V.K. FitzGerald. 2005. “The Standard of Living in Latin America during the Twentieth Century.” Economic History Review 54 (4): 765-796. [ Links ]

Astorga, Pablo, Ame Bergés & E.V.K. Fitz Gerald. 2011. “Productivity Growth in Latin America over the Long Run.” Review of Income and Wealth 57 (2): 203-223. [ Links ]

Barbero, María Inés & Fernando Rocchi. 2003. “Industry”. In A New Economic History of Argentina, edited by Gerardo della Paolera & Alan M. Taylor , 261-294. Cambridge: Cambridge University Press. [ Links ]

Bauer, Arnold. 2001. Good, Power, History and Latin America’s Material Culture. Cambridge: Cambridge University Press. [ Links ]

Beatty, Edward. 2001. Institutions and Investment: The Political Basis of Industrialization in Mexico before 1911. Stanford: Stanford University Press. [ Links ]

Benavente, José Miguel, Gustavo Crespi, Jorge Katz & Giovanni Stumpo. 1996. “Changes in the Industrial Development of Latin America.” Revista de la CEPAL (57): 49-72. [ Links ]

Bénétrix, Agustin S., Kevin H. O’Rourke & Jeffrey G. Williamson. 2012. “The Spread of Manufacturing to the Poor Periphery, 1870-2007.” NBER Working Paper, No. 18221. Cambridge: National Bureau of Economic Research. [ Links ]

Bértola, Luís & José Antonio Ocampo. 2012. The Economic Development of Latin America since Independence. Oxford: Oxford University Press. [ Links ]

Bielschowsky, Ricardo. 1998. “Evolución de las ideas de la CEPAL.” Revista de la CEPAL (RCEX01): 21-45. [ Links ]

Bielschowsky, Ricardo, ed. 2016. ECLAC Thinking: Selected Texts, 1948-1998. Santiago de Chile: ECLAC. [ Links ]

Bisang, Roberto, Mariana Fuchs & Bernardo Kosacoff. 1992. “Internacionalización de empresas industriales argentinas.” Desarrollo Económico 32 (127): 323-356. [ Links ]

Bortz, Jeffrey L. & Stephen Haber, eds. 2002. The Mexican Economy, 1870-1930: Essays on the Economic History of Institutions, Revolution and Growth. Stanford: Stanford University Press. [ Links ]

Brando, Carlos Andrés. 2012. “The Political Economy of Financing Late Development: Credit, Capital and Industrialisation; Colombia 1940-67”, PhD Dissertation, London School of Economics and Political Science. [ Links ]

Bresser-Pereira, Luiz Carlos. 2009. “Celso Furtado e a Teoria Econômica.” In Celso Furtado e a Formação Econômica do Brasil: edição comemorativa dos 50 anos de publicação (1959-2009), edited by Francisco da Silva Coelho & Rui Guilherme Granziera, 60-71. São Paulo: Editora Atlas. [ Links ]

Bulmer-Thomas, Victor. 1995. The Economic History of Latin America since Independence. Cambridge: Cambridge University Press. [ Links ]

Cárdenas, Enrique. 1987. La industrialización Mexicana durante la Gran Depresión. México: Colegio de México. [ Links ]

Cárdenas, Enrique. 1994. La hacienda pública y la política económica, 1929-58. México: Colegio de México. [ Links ]

Cárdenas, Enrique. 2000. “The Great Depression and Industrialization; the case of Mexico.” In An Economic History of Twentieth-Century Latin America, Volume 2: The Role of the Periphery in World Crisis, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 222-241. London: Palgrave/Macmillan. [ Links ]

Cardoso, Fernando Henrique & Enzo Faletto. 1979. Dependency and Development in Latin America. Berkeley: University of California Press. [ Links ]

Cerutti, Mário. 2000. Propietarios, empresarios y empresa en el norte de México. México: Siglo XXI. [ Links ]

Chang, Ha-Joon. 2003. Kicking Away the Ladder: Development Strategies in Historical Perspective. London: Anthem Press. [ Links ]

Chang, Ha-Joon. 1994. The Political Economy of Industrial Policy. London: Macmillan. [ Links ]

Coatsworth, John H. & Jeffrey Williamson. 2004. “Always Protectionist? Latin American Tariffs from Independence to the Great Depression.” Journal of Latin American Studies 36 (2): 205-232. [ Links ]

Dean, Warren. 1969. The Industrialization of São Paulo, 1880-1945. Austin: University of Texas Press. [ Links ]

Della Paolera, Gerardo, Xavier Durán Amorocho & Aldo Musacchio. 2015. “A Reassessment of Industrialization in South America: Argentina, Brazil, Chile and Colombia, 1890-2010.” Draft after the Conference “Industrialization in the global periphery, 1870-2008”. University of Oxford. [ Links ]

Díaz Alejandro, Carlos F. 1970. Essays on the Economic History of the Argentine Republic. New Haven: Yale University Press. [ Links ]

Díaz Alejandro, Carlos F. 1977. “Foreign Direct Investment by Latin Americans.” In Multinational from Small Countries, edited by Tamir Agmon & Charles P. Kindleberger, 167-195. Cambridge: Massachusetts Institute of Technology Press. [ Links ]

Díaz Alejandro, Carlos F. 2000. “Latin America in the 1930s.” In An Economic History of Twentieth-Century Latin America, Volume 2: The Role of the Periphery in World Crisis, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 17-49. London: Palgrave/Macmillan. [ Links ]

Díaz Fuentes, Daniel. 1994. Crisis y cambios estructurales en América Latina: México, Brasil y Argentina durante el período de entreguerras. México: Fondo de Cultura Económica. [ Links ]

Di Tella, Guido & Manuel Zymelman. 1967. Las etapas del desarrollo económico argentino. Buenos Aires: Editorial Universitaria de Buenos Aires-EUDEBA. [ Links ]

Draibe, Sônia. 1985. Rumos e matamorfoses: estado e industrialização no Brasil, 1930-1960. Rio de Janeiro: Paz e Terra. [ Links ]

Drinot, Paulo. 2000. “Peru, 1884-1930: a beggar siting on a bench of gold?” In An Economic History of Latin America in the Twentieth Century: Vol 1: The Export Age: The Latin American in the Late-nineteenth and Early-twentieth Centuries, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 152-187. London: Palgrave/St Antony’s. [ Links ]

Dye, Alan. 2006. “The Institutional Framework.” In The Cambridge Economic History of Latin America: Vol. II: The Long Twentieth Century, edited by Victor Bulmer-Thomas, John H. Coatsworth & Roberto Cortés Conde, 169-207. Cambridge: Cambridge University Press. [ Links ]

Economic Commission for Latin America (ECLA). 1967. The Latin American Economy in 1966. Excerpt from ECLA Survey. Santiago de Chile: United Nations Publications. [ Links ]

Evans, Peter. 1979. Dependent Development: The Alliance of Multinational, State and Local Capital in Brazil. Princeton: Princeton University Press. [ Links ]

Ffrench-Davis, Ricardo. 1973. Políticas económicas en Chile, 1952-70. Santiago de Chile: Ediciones Nueva Universidad. [ Links ]

FitzGerald, E.V.K. 1976. The State and Economic Development in Peru since 1968. Cambridge: Cambridge University Press. [ Links ]

FitzGerald, E.V.K. 2000. “Restructuring through the Depression: State and Capital Accumulation in Mexico, 1925-40.” In An Economic History of Twentieth-Century Latin America, Volume 2: The Role of the Periphery in World Crisis, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 212-232. London: Palgrave/Macmillan. [ Links ]

Furtado, Celso. 1959. Formação Econômica do Brasil. Rio de Janeiro: Fundo de Cultura. [ Links ]

Gerchunoff, Pablo & Lucas Llach. 2018. El ciclo de la ilusión y el desencanto: un siglo de políticas económicas argentinas de 1880 a nuestros días. Buenos Aires: Editorial Crítica. [ Links ]

Gereffi, Gary & Donald L. Wyman. 1990. Manufacturing Miracles: Paths of Industrialization in Latin America and East Asia. Princeton: Princeton University Press. [ Links ]

Gómez Galvarriato, Aurora. 1998. “The Evolution of Prices and real Wages in Mexico from the Porfiriato to the Revolution.” In Latin America and the World Economy since 1850, edited by John H. Coatsworth & Alan M. Taylor , 347-378. Cambridge: Harvard University/David Rockefeller Center for Latin American Studies. [ Links ]

Gómez Galvarriato, Aurora. 2002. “Measuring the Impact of Institutional Change in Capital-labor Relations in the Mexican Textile Industry, 1900-1930.” In The Mexican Economy, 1870-1930: Essays on the Economic History of Institutions, Revolution and Growth, edited by Jeffrey L. Bortz & Stephen Haber, 289-323. Stanford: Stanford University Press. [ Links ]

Grinberg, Nicolás. 2011. “Transformations in the Korean and Brazilian Processes of Capitalist Development between the mid-1950s and the mid-2000s: The Political Economy of Late Industrialisation”, PhD Dissertation, London School of Economics and Political Science. [ Links ]

Haber, Stephen. 2002. “Introduction: The Political Economy of Crony Capitalism.” In Crony Capitalism and Economic Growth in Latin America: Theory and Evidence, edited by Stephen Haber, xi-xxi. Stanford: Hoover Institution Press. [ Links ]

Haber, Stephen. 2006. “The Political Economy of Industrialization.” In The Cambridge Economic History of Latin America: Vol. II: The Long Twentieth Century, edited by Victor Bulmer-Thomas, John H. Coatsworth & Roberto Cortés Conde, 537-584. Cambridge: Cambridge University Press. [ Links ]

Haber, Stephen, Armando Razo & Noel Maurer. 2003. The Politics of Property Rights: Political Instability, Credible Commitments, and Economic Growth in Mexico, 1876-1929. Cambridge: Cambridge University Press. [ Links ]

Hamilton, Nora. 1982. The Limits of State Autonomy: Post-Revolutionary Mexico. Princeton: Princeton University Press. [ Links ]

Hewlett, Sylvia Ann & Richard S. Weinert, eds. 1982. Brazil and Mexico: Patterns in Late Development. Austin: Institute for the Study of Human Issues. [ Links ]

Hirschman, Albert O. 1963. Journeys Toward Progress: Studies in Economic Policy-Making in Latin America. New York: Twentieth Century Fund. [ Links ]

Hirschman, Albert O. 1981. Essays in Trespassing. Economics to Politics and Beyond. Cambridge: Cambridge University Press. [ Links ]

Hofman, André A. 2000. The Economic Development of Latin America in the Twentieth Century. London: Edward Elgar. [ Links ]

Jaramillo Echeverri, Juliana, Adolfo Meisel-Roda & María Teresa Ramírez-Giraldo. 2015. “The Great Depression in Colombia: A Stimulus to Industrialization, 1930-1953.” Borradores de Economía, No. 892. Bogotá: Banco de la República. [ Links ]

Jenkins, Rhys O. 1987. Transnational Corporations and Uneven Development: The Internationalization of Capital and the Third World. London: Methuen. [ Links ]

Jenkins, Rhys O. 1991. “The Political Economy of Industrialization: A Comparison of Latin American and East Asian Newly Industrializing Economies.” Development and Change 22 (2): 197-232. [ Links ]

Katz, Jorge & Bernardo Kosacoff. 1983. “Multinationals from Argentina.” In The New Multinationals: The Spread of Third World Enterprises, edited by Sanjaya Lall, 137-219. London: Wiley. [ Links ]

Kaufman, Robert R. 1985. The Politics of Debt in Argentina, Brazil and Mexico: Economic Stabilization in the 1980s. Berkeley: University of California Press. [ Links ]

Kaufman, Robert R. 1990. “Stabilization and Adjustment in Argentina, Brazil and Mexico.” In Economic Crises and Policy Choice: The Politics of Adjustment in the Third World, edited by Joan Nelson, 63-112. Princeton: Princeton University Press. [ Links ]

Kay, Cristobal. 1991. “Reflections on the Latin American Contribution to Development Theory.” Development and Change (22): 31-68. [ Links ]

Kay, Cristobal. 2011. Latin American Theories of Development and Underdevelopment. London: Routledge. [ Links ]

Leopoldi, Maria Antonieta. 1994. Política e Interesses: As Associações Industriais e Política Econômica e o Estado na Industrialização Brasileira. São Paulo: Paz e Terra. [ Links ]

Lewis, Paul H. 1990. The Crisis of Argentine Capitalism. Durham: University of North Carolina Press. [ Links ]

Love, Joseph L. 1994. “Economic Ideas and Ideologies in Latin America since 1930.” In The Cambridge History of Latin America: Vol. VI, Pt. 1: 1930 to the Present, edited by Leslie Bethell, 393-460. Cambridge: Cambridge University Press. [ Links ]

Love, Joseph L. 2005. “The Rise and Decline of Economic Structuralism in Latin America.” Latin American Research Review 40 (3): 100-125. [ Links ]

Maddison, Angus. 1991. “Economic and Social Conditions in Latin America, 1913-1950.” In Long-Term Trends in Latin American Economic Development, edited by Miguel Urrutia, 1-22. Baltimore: Johns Hopkins University Press/Inter-American Development Bank. [ Links ]

Marichal, Carlos & Mário Cerutti, eds. 1997. Historia de las grandes empresas en México, 1850-1930. México: Universidad Autónoma de Nuevo León/Fondo de Cultura Económica. [ Links ]

McGreevy, William P. 1971. An Economic History of Colombia, 1845-1930. Cambridge: Cambridge University Press. [ Links ]

Mesa Lago, Carmelo. 1991. Social Security and Prospects for Equity in Latin America. Washington D.C.: World Bank. [ Links ]

Mesa Lago, Carmelo. 2008. Reassembling Social Security: A Survey of Pension and Healthcare Reform in Latin America. Oxford: Oxford University Press/Pan-American Health Organization. [ Links ]

Mesquita Moreira, Mauricio. 1995. Industrialization, Trade and Market Failures: The Role of Government Intervention in Brazil and South Korea. London: Macmillan. [ Links ]

Montevideo-Oxford Latin American Economic History Database (MOxLAD). http://moxlad-staging.herokuapp.com/home/enLinks ]

Naya, Seiji, Miguel Urrutia, Shelley Mark & Alfredo Fuentes. 1989. Lessons in Development: A Comparative Study of Asia and Latin America. San Francisco: International Center for Economic Growth. [ Links ]

North, Douglass C., William Summerhill & Barry R. Weingast. 2000. “Order, Disorder and Economic Change: Latin America vs. North America.” In Governing for Prosperity, edited by Bruce Buenos de Mesquita & Hilton Roots, 17-58. New Haven: Yale University Press. [ Links ]

Ocampo, José Antonio. 2000. “The Colombian Economy in the 1930.” In An Economic History of Twentieth-Century Latin America, Volume 2: The Role of the Periphery in World Crisis, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 117-143. London: Palgrave/Macmillan. [ Links ]

Ocampo, José Antonio. 2007. “La economía colombiana en la década del treinta.” In Crisis Mundial, Protección e Industrialización, edited by José Antonio Ocampo & Santiago Montenegro, 19-56. Bogotá: CEREC y FESOL/Editorial Norma. [ Links ]

Ocampo, José Antonio & Jaime Ros. 2011. “Shifting Paradigms in Latin American Economic Development.” In The Oxford Handbook of Latin American Economics, edited by José Antonio Ocampo and Jaime Ros, 3-25. Oxford: Oxford University Press. [ Links ]

O´Donnell, Guillermo. 1973. Modernization and Bureaucratic Authoritarianism in South American Politics. Berkeley: University of California Press. [ Links ]

O´Donnell, Guillermo. 1999. Counterpoints: Selected Essays on Authoritarianism and Democratization. Notre Dame: University of Notre Dame Press. [ Links ]

Ospina Vásquez, Luis. 1955. Industria y protección en Colombia 1810-1930. Medellín: ESF. [ Links ]

Ospina Vásquez, Luis. 1963. El plan agrícola. Medellín: Gran América. [ Links ]

Palacios, Marco. 1980. Coffee in Colombia: An Economic, Social and Political History, 1850-1970. Cambridge: Cambridge University Press. [ Links ]

Palma, Gabriel. 2000. “From an Export-Led to an Import-Substituting Economy: Chile 1914-39 Gabriel Palma.” In Latin America in the 1930s: The Role of the Periphery in World Crisis, edited by Rosemary Thorp, 50-80. London: Palgrave/St Antony’s. [ Links ]

Palma, Gabriel. 2015. “External Disequilibria and Internal Industrialisation: Chile 1914-35.” In Latin America, Economic Imperialism and the State, edited by Christopher Abel & Colin M. Lewis, 318-338. London: Athlone/Bloomsbury. [ Links ]

Pineda, Yovanna. 2009. The Industrialization of Argentina, 1890-1930. Stanford: Stanford University Press. [ Links ]

Prados de la Escosura, Leandro. 2007. “When did Latin America Fall Behind? Evidence from Long Run International Inequality.” In The Decline of Latin America Economies: Growth, Institutions and Crisis, edited by Sebastian Edwards, Gerrardo Esquivel & Graciela Márquez, 15-57. Chicago: University of Chicago Press. [ Links ]

Prados de la Escosura, Leandro. 2015. “Human Development as Positive Freedom: Latin America in Historical Perspective.” Journal of Human Development and Capabilities 16 (3): 342-373. [ Links ]

Prebisch, Raúl. 1950. The Economic Development of Latin America and its Principal Problems. New York: United Nations. [ Links ]

Rapoport, Mário. 2012. Historia económica, política y social de la Argentina, 1880-2003. Buenos Aires. [ Links ]

Rivarola Puntigliano, Andrés & Örjan Appelqvist. 2011. “Prebisch and Myrdal: Development Economics in the Core and on the Periphery.” Journal of Global History 6: 29-52. [ Links ]

Rocchi, Fernando. 2006. Chimneys in the Desert: Industrialization in Argentina during the Export Boom Years, 1870-1930. Stanford: Stanford University Press. [ Links ]

Rossoto Ioris, Antonio Augusto & Rafael Rossoto Ioris. 2013. “Assessing Development and the Idea of development in the 1950s in Brazil.” Revista de Economía Política 33 (3): 411-426. [ Links ]

Salvucci, Richard. 2006. “Export-led Industrialisation.” In The Cambridge Economic History of Latin America: Vol. II: The Long Twentieth Century, edited by Victor Bulmer-Thomas, John H. Coatsworth & Roberto Cortés Conde, 249-292. Cambridge: Cambridge University Press. [ Links ]

Schneider, Ben Ross. 2004. Business Politics and the State in 20th Century Latin America. Cambridge: Cambridge University Press. [ Links ]

Serra, Jose. 1979. “Three Mistaken Theses Regarding the Connection between Industrialization and Authoritarian regimes.” In The New Authoritarianism in Latin America, edited by David Collier, 99-163. Princeton: Princeton University Press. [ Links ]

Sikkink, Kathryn. 1991. Ideas and Institutions: Developmentalism in Brazil and Argentina. Ithaca: Cornell University Press. [ Links ]

Solís, Leopoldo. 1970. La realidad econónomica mexicana. México: Siglo XXI. [ Links ]

Sunkel, Osvaldo. 1993. “Introduction: In Search of Development Lost.” In Development From Within: Towards a Neostructuralist Approach for Latin America, edited by Osvaldo Sunkel, 1-4. Boulder: Lynne Rienner. [ Links ]

Suzigan, Wilson. 1986. Indústria brasileira: Origem e Desenvolvimimento. São Paulo: Editora Brasiliense. [ Links ]

Suzigan, Wilson & Annibal Villela. 1997. Industrial Policy in Brazil. Campinas: UNICAMP/IE. [ Links ]

Taylor, Alan. 1998. “On the Costs of Inward-Looking Development: Price Distortions, Growth, and Divergence in Latin America.” The Journal of Economic History 58 (1): 1-28. [ Links ]

Thorp, Rosemary. 2000. “Introduction.” In An Economic History of Twentieth-Century Latin America, Volume 2: The Role of the Periphery in World Crisis, edited by Enrique Cárdenas, José Antonio Ocampo & Rosemary Thorp, 1-16. London: Palgrave/Macmillan. [ Links ]

Thorp, Rosemary.1991. Economic Management and Economic Development in Peru and Colombia. Basingstoke: Macmillan. [ Links ]

Thorp, Rosemary. 1998. Progress, Poverty and Exclusion: an economic history of Latin America in the 20th century. New York: Inter-American Development Bank. [ Links ]

Thorp, Rosemary & Geoffrey Bertram. 1978. Peru, 1890-1977: Growth and Policy in an Open Economy. London: Palgrave/Macmillan. [ Links ]

Urrutia, Miguel, ed. 1991. Long Term Trends in Latin American Economic Development. Baltimore: Johns Hopkins University Press/Inter-American Development Bank. [ Links ]

Versiani, Flávio R. 1979. “Industrial Investment in an ‘Export Economy’: The Brazilian Experience before 1914.” Working Paper Institute of Latin American Studies, No. 2. London: University of London. [ Links ]

Versiani, Flávio R. 1987. A década de 20 na industrialização brasileira. Rio de Janeiro: Instituto de Planejamento Econômico e Social/Programa Nacional de Pesquisa Econômica. [ Links ]

Villela, Annibal V. & Wilson Suzigan. 1973. Política do governo e crescimento da economia brasileira, 1889-1945. Rio de Janeiro: IPEA. [ Links ]

Williamson, Jeffrey G. 2011. Trade and Poverty: When the Third World Fell Behind. Cambridge: Massachusetts Institute of Technology Press. [ Links ]

Wirth, John D. 1970. The Politics of Brazilian Development, 1930-45. Stanford: Stanford University Press. [ Links ]

How to cite: Lewis, Colin M. 2019. “CEPAL and ISI: Reconsidering the Debates, Policies and Outcomes”. Revista de Estudios Sociales 68: 8-26.https://doi.org/10.7440/res68.2019.02

* Elements of this article are based on a larger, independent research project part-funded by the LSE Staff Research Fund, with incidental assistance from the Brazilian Association of Economic History (ABPHE).

1The literature is vast: for accessible, critical introductory essays to structuralism and its dissonances see Kay 1991 & 2011; Jenkins 1991; Love 2005; Ocampo & Ros 2011.

2Formed in 1948 as the United Nations Organisation Economic Commission for Latin America, the designation and scope of the Commission was changed in 1984 to the Economic Commission for Latin America and the Caribbean, in part reflecting the independence of former colonial territories, many whom had previously been associate members.

3The original member countries were Bolivia, Chile, Colombia, Ecuador and Peru; Venezuela became a member in 1973.

4Land reform involved little more that the allocation of titles of ownership to campesinos; in addition to land distribution, agrarian reform entailed infrastructure investment in the countryside, the provision of scientific and commercial information about products and markets and credit and capital for producers, and generalized support for rural entrepreneurs. The latter was often beyond the capacity —or the ability— of the state.

Received: June 01, 2018; Accepted: October 29, 2018

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License