From Populism to Neoliberalism: The Political Economy of Latin American Import-Substitution Industrialization: Argentina,Brazil,Mexico and Colombia in Comparative Perspective

Argentina: from Peronism to Dictatorship

Argentina’s experience was paradigmatic of that of national economies structured around the production of primary commodities and capital’s recovery of ground rent through import-substitution industrialization (see Iñigo-Carrrera, 2006; 2007). The reasons for this are found in the historical and natural conditions under which this modality of capital accumulation developed there. First, despite playing a subsidiary role in the colonial economy, Argentina’s Humid Pampas became Latin America’s first region to enter a postindependence high-growth phase based on primary-commodity exports because of its favorable climatic and soil conditions for producing first wool and later meat and cereals. Second, this meant that capital had access to particularly abundant ground rent materialized in working-class means of consumption partly produced by small-size agrarian capitals. Third, given the production requirements for cereals and dairy, the process involved the early import of petty capitalists and wage laborers from the Southern European regions that were being outcompeted by Argentine production. These conditions produced comparatively large domestic markets because manufacturing capital’s appropriation of ground rent came to be mediated by working-class consumption of low-priced foodstuffs and by agrarian capital’s consumption of high-priced means of production. Fourth, the high-rent content of Argentina’s agrarian production resulted in a particularly high level of net “taxation” (explicit or otherwise) of rent-bearing commodities and, ceteris paribus, in relatively low investment, intensive or extensive, in agrarian land. Regionwide, policy-induced restrictions on agrarian development were particularly severe, limiting the positive impact of intensive investment on the volatility inherent in climate-influenced agrarian production and resulting in a particularly fluctuating, slow-growing ground rent (Figure 3).

OPEN IN VIEWER

Figure 3.

Industrial value added in 2005 US$, 1947–2010 (Argentina, Iñigo-Carrera. 2007; Brazil, Grinberg, 2016; Colombia and Mexico, MOxLAD, n.d.; Mitchell, 2013).

Though Argentine industrialization accelerated during the first decades of the twentieth century as a by-product of primary-exports growth and international trade disruptions, the end of World War II marked the beginning of profound transformations in the economic forms of capital accumulation. With substantially increased ground rent, industrialization developed not only quantitatively but also qualitatively (see Figures 2 and 3). Yet, for industrial capital to move into durable- and, later, capital-goods production, more than infant-industry protection and subsidization was required; a supporting basic-inputs industry was also needed. Given their relative underdevelopment and the complexity of the production processes involved, industrial “deepening” could take place only through state mediation in the centralization of industrial and moneylending capital.

The rapid development of state-owned enterprises in scale-intensive heavy industry and public utilities to support small-scale private-capital accumulation came about through the actions of a government backed by the national bourgeoisie and, especially, by increasingly state-controlled working-class organizations, and so did the advanced payment of state’s high-interest foreign debts and the nationalization of public-utility companies on onerous terms (with ground rent in the form of bulky credits and reserves accumulated during the military conflict). Despite their limited character, these policies appeared in economic theories and public discourses as being necessary to promote independent national development through industrialization and social justice since they were combined with state-managed increases in direct and indirect wages that enlarged the urban workforce through internal migration while rapidly upskilling it to suit the requirements of expanding industrial capital (Figure 4). These state forms mediating Argentina’s economic development came about through a political process known as nationalistic populism, managed by the newly formed, unions-led, broad-based and reform-oriented Peronist (later Justicialist) Party presided over by the military strongman Juan Perón; this sidelined labor’s preexisting class-based industrial and political organizations.

OPEN IN VIEWER

Figure 4.

Purchasing power of annual industrial wages in 2005 US$, 1939–2015 (Argentina, Iñigo-Carrera, 2007; Brazil, Grinberg, 2016; Colombia, DNP, n.d.; MOxLAD, n.d. Mitchell, 2013; Mexico, Bortz, 1976; INEGI, n.d.). Local-currency values were transformed into US$ using the World Bank’s purchasing-parity exchange rates and indexed using the growth rate of real wages.

By the early 1950s, when the U.S. and European economies were fully stabilized and the Korean War commodities boom had waned, material conditions in the Argentine domestic market were ripe for transnational capital to accelerate its entrance into the durable-goods sector. First, there was enough productive capacity in state-owned basic-input and heavy-engineering industry and in a broad range of public utilities. As noted, the ground-rent materialized in these means of production was transferred to private industrial capital through underpriced sales and overpriced purchases. Second, there was a large, trained industrial workforce. Third, there was an extensive network of private domestic small capitals able to supply and service those industries, transferring a portion of their profits in the process. Fourth, because of these developments and the nature of Argentina’s agrarian production, there was a large market for manufactures, especially white goods and motor vehicles. Finally, the ground rent materialized in exports and domestic consumption, though diminished, was sufficiently large to subsidize capital imports, profit remittances, and labor power for industrial capital.

By eroding the Perón government’s material bases of support and leading to its downfall, the mid-1950s economic crisis, resulting from successive droughts and falling international primary-commodity prices, accelerated the process; it had been long in the making. Once the 1955–1958 military dictatorship had removed the residual political, legal, and ideological restrictions to foreign-capital inflows, a pseudo-democratic government led by the pre-Peronist broad-based, petty-bourgeoisie-led party was elected to manage open-gate manufacturing-capital inflows in the name of modernization and development instead of autarky and self-determination; it had the informal support of the proscribed union-based Peronism. Nationalistic populism thus gave way to developmental populism as the political form of capital accumulation; the former’s historical potentiality had been to prepare the ground for the latter. Economic theories now recommended foreign investment in manufacturing to accelerate technical change and overcome balance-of-payment restrictions to industrial development. The structuralism of the Economic Commission for Latin America and the Caribbean became the general theoretic-ideological expression of a process of capital accumulation revolving around primary-commodity production and the appropriation of ground rent by different social subjects, thereafter especially foreign-origin industrial capital. Along with those already there, capitals entering the Argentine economy for investment in manufacturing did so with a scale of production reduced to the size of the internationally small, increasingly protected domestic market, where they would sell their entire output while recycling outdated means of production.

The large inflow of manufacturing capital, replacing capital invested in public utilities and state debts as landowners’ main partners in the business of ground-rent appropriation, took the form of increasing industrialization and the further internationalization of the national economy. Low-priced purchases of national capitals by foreign investors constituted a form of appropriation by the latter of the ground rent materialized in their fixed assets. A process of centralization thus took place during the 1960s, accelerated by the mid-decade economic crisis, that also manifested itself in a state-managed wage squeeze to generate extraordinary surplus value and complement a contracting ground rent in sustaining the valorization of capital (see Figures 2 and 4). As does any other change in the economic forms of capital accumulation, this one came about through a political reshaping—a core-state-supported bureaucratic-authoritarian regime. Repression-backed economic liberalization began to complement developmentalism as the political form of capital accumulation through ground-rent recovery in Argentina and U.S.-fed anticommunism as its ideological cover.

Though economic growth accelerated in the late 1960s and early 1970s, especially when international primary-commodity markets boomed, the contradictions inherent in this national form of global-capital accumulation soon became a crisis. The expansion of ground rent of the 1970s manifested itself in further state-led industrialization driven by the increasingly nationalistic-populist politics and independent geopolitical strategies of the relegalized Justicialist Party, but it was clearly unsustainable and short-lived. After peaking in late 1973, agrarian prices started a three-decade-long decline that negatively affected the ground rent available for appropriation (see Figure 2).

With ground rent contracting and capital’s requirements expanding as local production costs continued to lag behind world-market norms and especially as electronic-based automation advanced and disciplined East Asian labor power became suitable for standardized production, Argentina’s industrialization came up against its structural limit. Thereafter capital drew on still more contradictory and precarious sources of extraordinary surplus value to support its limited valorization: a portion of the value of labor power, foreign loans, and privatization funds. In the short run, the latter two supplemented ground rent; in the long run, they acted as a form of ground-rent appropriation through interest-rate differentials and profits. State-mediated import-substitution industrialization was reproduced in increasingly limited neoliberal forms, including the partial displacement of manufacturing capital from the business of ground-rent appropriation through state-asset privatization and trade opening and the extension of narrow domestic markets through regional integration. Though these transformations became consolidated under post-mid-1980s popular-based democratic governments, they began by means of a military dictatorship that violently repressed the working class to drive wages down in the wake of a severe industrial crisis.

Changes in politico-economic dynamics came about, as usual, through particular ideological expressions. As ground rent became incapable of sustaining the previous scale of accumulation in the industrial sector and policies mediating its appropriation by industrial capital gave way to structural reforms supported by core-country governments and global-governance institutions, structuralist developmentalism was increasingly replaced by new-institutional-economics-informed neoliberalism as the general theoretic-ideological form of capital accumulation structured around the appropriation of ground rent by different social subjects, especially its recovery by rent-paying international capital.

Brazil: From New State to New Unionism

As in Argentina, import-substitution industrialization in Brazil was the economic form of capital accumulation structured around the production of primary commodities for world markets and the appropriation of ground rent during a period when rent-paying manufacturing capital, increasingly of foreign origin, became landowners’ main “business partner” (see Grinberg, 2016, for a more extensive analysis). Differences between the Brazilian and the Argentine experience arose from the conditions under which this process was realized. First, during the spontaneous (1920s–1930s) and early state-led (1940s–1950s) stages of import-substitution industrialization, the total ground rent available for appropriation in Brazil was not only less than in Argentina but also largely contained in the prices of exported commodities that did not constitute the bulk of working-class consumption—coffee and cacao. This natural condition resulted not only in less ground rent for capital to appropriate through small-scale industrial production but also in different economic policies and political developments mediating the process. Second, in contrast to Argentina, where capital imported most of its initial stock of industrial and agrarian labor, Brazil already had a large surplus population inherited from the pre-Republican periods that could be used for simple industrial activities, urban services, and, crucially, low-skilled tropical arboriculture. This historical condition allowed capital to differentiate the reproduction of the working class to a much greater extent than in Argentina, thus generating a permanent flow of extraordinary surplus value that both expanded (through low rural wages) and complemented (through low urban wages) the ground rent available for appropriation (see Figure 4). Third, because of the lower rent content, policy-set restrictions on capital investments in the agrarian sector were milder; as the period progressed, Brazilian ground rent came from increasingly diverse sources in terms of commodities and regions, resulting in a bulkier and more stable flow than in Argentina (see Figure 2). Because of these initial conditions and their dynamics, import-substitution industrialization in Brazil went beyond that in Argentina and produced less organized social conflict despite the greater incidence and entrenchment of poverty (see Figure 3).

Though primary-commodity production and raw-material processing for world markets began in Brazil earlier than in Argentina, by the early twentieth century its most prosperous coffee-growing region, São Paulo, lagged behind Argentina’s Humid Pampas in terms of value produced, rent content, technical complexity, and productive linkages (Maddison, 1993: 33–35; Skidmore, 2007: 3–12). Nevertheless, as in Argentina, post-Depression Brazil enjoyed a rapid industrial recovery, especially in terms of world-market trends. With slow-growing export earnings and therefore limited import power, production for domestic markets expanded on installed capacity. Again as in Argentina, the reduced ground rent was shared by landowners, foreign-owned public-utility companies, and, through continuous renegotiations, foreign creditors (Diaz Fuentes, 1994: 127, 140–142, 166–170). Brazilian incipient postcrisis industrial development, however, required not only political control over skilled workers to limit wage recovery, as in Argentina, but also state regulation of capital’s consumption of unskilled labor power; crisis-induced rural-urban migration led to an oversupply of labor, thus undermining labor’s already precarious reproduction. Several labor protection laws were passed (somewhat later than in Argentina) and merged into the overarching 1943 Labor Code, which was contemporary to Argentina’s development in that direction and regulated capital-labor relations thereafter. Moreover, industrial development required large state investments in lagging infrastructure and basic-industry production, and, in contrast to the situation in rent-rich Argentina, this called for financial and technical support from core-country states in exchange for Brazil’s World War II participation. These processes came about through increasingly corporatist institutions that culminated in the formation of the centralized and authoritarian New State (1937–1945), led by the strongman Getúlio Vargas, whose main political backing originated in the domestic-markets-oriented (industrial and agrarian) bourgeoisie (Skidmore, 2007: 12–47).

As in Argentina, the global-market conditions prevailing at the end of World War II accelerated previous developments. With expanded ground rent in the form of accumulated war credits and growing primary-commodity exports, capital accumulation in Brazil’s industrial sector increased. Private capitals renewed their equipment through state-managed exchange-rate overvaluation, while state-owned enterprises in basic industry, developmental banks, and public utilities were created either from scratch or through the nationalization, at inflated prices, of foreign-owned assets. Industrial-sector wages increased steadily as the demand for semiskilled labor power sharply expanded (see Figures 1–4), and so did rent-funded, state-controlled public-sector and union expenditures on the reproduction of labor. As in Argentina, these transformations came about through an interclass alliance and found politico-ideological expression in nationalistic populism. Yet, lacking the potential of the contemporary Argentine developments (resulting from the quantity and quality of the ground rent available for appropriation and their expression in stronger industrial development and a more homogeneous workforce), a two-partner coalition government between state-created parties representing the national bourgeoisie and the urban working class (rather than a labor-led national-front government) politically mediated the transformations in the Brazilian economy. Resulting in reduced working-class power, the extra flexibility afforded by the Brazilian arrangements allowed the coalition to express politically the large foreign-capital inflows into manufacturing that ensued through the mid-1950s, when the populist state became developmentalist (1956–1964). Not only was the institutional break mediating the transition from nationalistic to developmental populism less violent than in Argentina (involving only the president’s suicide and a pseudo-coup d’état rather than a full-scale one that included airborne bombing of the government’s headquarters) but the working-class party was able to take part in it actively and officially. This time, contrary to the situation in 1950–1955, the interclass alliance running the state was led by the bourgeoisie’s political organization.

Despite Brazil’s more stable political and institutional development, the contradictions inherent in this form of capitalist development appeared there sooner than in Argentina. Already in the early 1960s, the slowly increasing ground rent proved incapable of sustaining the expanded reproduction of capital accumulation through import-substitution industrialization: a new balance-of-payments crisis emerged as slow-growing export earnings became insufficient to fund fast-growing capital-goods imports and profit remittances; inflation accelerated as public-sector finances deteriorated while ground-rent taxation decreased; unemployment mounted as job creation failed to keep pace with rural-urban migration. These economic dynamics took shape at a peak of social conflict and found ideological expression in dependency theory, a product of structuralist developmentalism that spread across the region with an intensity reflecting that of these contradictions and the potential for their being partly overcome. In its radical versions, dependency theory pointed to the need for a revolutionary solution to the problem of underdevelopment (Marini, 1972), in its reformist versions to the low labor-intensity of imported technologies and consumption patterns (Cardoso, 1973; Furtado 2020 [1973]). In all its versions, explicitly or otherwise, it pointed to the lack of an advanced capital-goods industry as the main cause of Latin America’s dependency and the structural constraint to be overcome by direct state regulation/control.

The two-party interclass alliance giving political form to the populist-democratic period (1946–1964) in both its nationalistic and its developmental phases could not, however, mediate the economic changes taking place in the second half of the 1960s, when the centralization of capital that followed the 1963–1966 crisis eliminated, as in Argentina, a large proportion of small nationally owned manufacturing companies. Nor could it achieve a squeeze on manual workers’ wages as ground rent became increasingly insufficient to sustain capital’s valorization on an expanded scale. As in Argentina, these changes came about through a politically repressive military regime that instituted an economically liberal program (1964–1967). On these reformed bases and given the large late-1960s Eurodollar inflows that they facilitated to supplement the stagnant ground rent, the economy entered a new phase of industrialization bolstered by lower wages and state investment in debt-funded heavy-industry and infrastructural projects; thus a “miraculous” growth process ensued (1968–1973). Portraying it as “associated-dependent development,” dependency theory now fed discourses shaping Brazilian industrialization. Its reformist branch became mainstream, while military-led economic policy turned increasingly dirigiste and the so-called triple alliance (between foreign, state, and national capital) gave economic expression to capital accumulation at this stage of its historical development.

Like the populist-democratic institutional arrangements, the military regime offered Brazil, in contrast to “pendular” Argentina, the flexibility necessary for the political expression of the period’s sharply but erratically increasing ground rent, since it did not initially depend on across-the-board wage increases. Provided with a large surplus population, capital could postpone these until it was no longer feasible to reproduce an industrial workforce with the productive attributes necessary to sustain its accumulation. Only in the late 1970s did manufacturing wages increase sharply as state-led industrialization took a great leap forward into heavy industry and capital-goods production while the commodities booms manifested themselves in expanding ground rent and lendable-capital inflows. The wage increases that resulted from capital’s expanded demand for skilled labor power peaked through a short-lived upsurge in industrial actions and broader civil-society movements led by a new brand of politically independent unionism. Thus, as an expression of these accumulation dynamics, in 1973 the military in control of the state placed in power a nationalistic leadership that implemented a developmentalist economic program (in 1974), an independent foreign policy (in 1977), and a slow-paced political opening (1974–1979) that together realized the deepening of manufacturing capital mediating the appropration of the expanding ground rent and complementary loanable-capital inflows.

The structural limits of import-substitution-industrialization-based accumulation created by the relationship between manufacturing capital’s growing requirements for ground rent and the amount available for appropriation were reached in Brazil a few years later than in Argentina. An increase in the production of rent-bearing commodities and a decline in the rent-dependence of manufacturing capital drove the crisis of state-led import-substitution industrialization until 1980. Nevertheless, the economic and political processes through which the subsequent transformations came about were similar, parallel, and interrelated.

Mexico: From Revolution to Shared Development

In general terms, Mexican import-substitution industrialization expressed similar accumulation dynamics as those of Argentina and Brazil. Its singularities sprang not only from the relatively limited ground rent available for appropriation before the early-1970s offshore oil discoveries but also from two qualitative differences. First, Mexican ground rent came largely from metal and oil extraction rather than agrarian production. Since mining is less affected by weather conditions, the flow of ground rent to the Mexican economy was more stable, and since mining lands (including oil resources) are less dispersed geographically, public ownership of them could be asserted, leading to more state-intensive forms of ground-rent recovery by industrial-sector capital. Second, relatively poor natural conditions for agrarian production meant that capital’s efforts to increase ground rent to sustain import-substitution industrialization required more extensive state mediation than elsewhere in support of agrarian capital and petty-commodity producers, further stabilizing the flow of ground-rent available for appropriation.

As did Argentina’s and Brazil’s, Mexican industrialization gained momentum in the late nineteenth century as a by-product of primary-exports expansion (Knight, 2018: 244–248). The surging economic power of domestic-markets-oriented capitalists and urban workers also contributed there to the removal of an oligarchic regime. In Mexico, however, this transformation also involved interregional, international, and intersectoral rebalancing—the growing importance of northern Mexican capital in the national economy and the consolidation of rent-paying U.S. manufacturing capital as a leading partner in the business of ground-rent appropriation— and the partial elimination of a population far exceeding capital’s valorization requirements. These processes came about, however, not through revolt-led electoral politics (as in mid-1910s Argentina) or a bloodless bourgeois revolution (as in early-1930s Brazil) but through a lengthy and violent U.S.-supported popular uprising (Womack Jr., 1986).

As in Argentina and Brazil, the consolidation of manufacturing capital required changes in state dynamics and therefore in the politico-institutional arrangements mediating the accumulation process. These changes took shape in policies accelerating the centralization of capital under public ownership in “natural monopolies” and industrial-worker upskilling, which was especially important in view of the limited capacity of the Mexican economy to attract ready-to-use foreign labor. They also required the reorganization of agrarian production to increase the supply of low-priced foodstuffs for the growing industrial workforce (Meyer, 1986).

Initially, this process took shape in an agrarian reform that partly reversed the previous state-led privatization of communal lands, leading to community-based productive fragmentation in nonirrigated areas and to cooperative-based centralization of scale-intensive irrigated production. This stage peaked during Cárdenas’s nationalistic-populist government (1934–1940), when capital accumulation was recovering from the Great Depression depths and demand for food and inputs was accelerating (Hewitt de Alcántara, 1999: 21–30). By eliminating landowners, the agrarian reform increased the portion of intramarginal-land surpluses available for appropriation while expanding the area of extramarginal semiarid land under cultivation (Sonnenfeld, 1992).

By the early 1940s, however, the limited potentiality of the agrarian-reform program to expand ground rent vis-à-vis the emerging technologies of large-scale intensive agriculture was self-evident to those personifying the political representation of capital accumulation, and the program began to be reversed. Not only did land redistribution decelerate markedly, to focus thereafter on low-quality plots, but size limits were lifted for new allocations while already distributed lands began to be rented out to commercial producers. The small scale of the postreform landholders and the illegality of operations involving community-owned land meant that payments received by peasants-turned-rentiers were smaller than they would have been under normal market conditions (Hewitt de Alcántara, 1999: 191–195). Thus, by further reducing landowner power (i.e., increasing the proportion of the market price pocketed by agrarian capital), the restrictions that landed property generally imposes on intensive capital investment in land were further reduced (Marx, 1981: 882–907). Intensive agriculture, nevertheless, required not only land consolidation but also investment in irrigation and research and the centralized, often subsidized provision of seeds, agrochemicals, and fuel. Only by combining these inputs could agrarian production in the northern alluvial plains, where most resources were concentrated, increase sufficiently to yield normal profits and some rent. The scientific and technical research behind the so-called Green Revolution was largely conducted and funded by U.S. public and private institutions, so extended was U.S. capital’s interest in ground-rent appropriation in Mexico (Hewitt de Alcántara, 1999: 56–88, 99–111).

The Mexican state not only was central in increasing the ground rent available for appropriation through agrarian policies but also played a key part in transferring ground rent to manufacturing capital through industrial-promotion policies such as those of Argentina and Brazil. Controlling the most dynamic branches, foreign-origin companies would pocket, as elsewhere, the bulk of these resources (Newfarmer and Mueller, 1975; Fitzgerald, 1985: 212–216; Maddison, 1993: 182–187, 207–210), but the economic forms of this process diverged somewhat from those prevailing in Argentina and Brazil. First, since the bulk of the ground rent accrued to mining lands, sector-specific taxes funding state budgets prevailed over exchange-rate overvaluation directing ground rent to private capital (see Figure 1). The revolutionary constitution renationalized subsoil resources, strengthening the state’s power to tax rent-bearing metal-mining commodities produced by private, mostly international companies. Because of its extended productive linkages and its increasing domestic-market orientation, the oil industry was fully nationalized in 1938, as it had been in Argentina and would be in Brazil, with foreign owners receiving full compensation for their assets (Maurer, 2011). Thereafter, oil rents were transformed into profits for industrial capital through state-company activities (i.e., low-priced energy sales and high-priced purchases of inputs and fiscal contributions [Morales, 1992]). Second, in contrast to what happened in Argentina and Brazil, the state-mediated appropriation of agrarian rent by industrial capital involved the management of primary-commodity domestic trade to complement (mild) exchange-rate overvaluation or, eventually, to compensate agrarian capitals for its negative impact on farm-gate prices (Maddison, 1993: 187–204; Hewitt de Alcántara, 1999: 56–98). Qualitative differences notwithstanding, with less ground rent available for appropriation than in Argentina and Brazil, import-substitution industrialization was shallower (see Figures 2 and 3). State investment in basic-input production and private-sector equipment manufacturing was less extensive (Knight, 2018: 258), and therefore foreign-controlled durable-consumer- and capital-goods industries developed less intensively (Newfarmer and Muller, 1975; Maddison, 1993).

As the economic forms of capitalist development, Mexican political mediations during state-led import-substitution industrialization also developed singularities as an expression of the distinctive conditions under which this national modality of global-capital accumulation was realized. In general terms, the limited potentiality of the available ground rent to expand and deepen import-substitution industrialization produced a collective worker with simpler productive attributes than in Argentina and Brazil and less extensive state management of industrial capital, both of them normally associated with the agency of working-class organizations. As a result, their political power was even more limited than in Brazil, let alone Argentina. Moreover, in contrast to the situation in those countries, the state-led development of peasant agriculture meant that rural workers were also included in the political representation of capital accumulation, further diluting industrial workers’ autonomy. This relative weakness meant that capital required only mild state repression to undertake wage compression/restraint through labor-market segmentation. In specific terms, the lower volatility of the flow of ground rent because of the significance of mining and the greater capital-intensiveness of agrarian production meant that, in contrast to the Argentine and Brazilian experiences, changes in state policies mediating the Mexican process of capital accumulation through state-led import-substitution industrialization were the result of less antagonistic changes in its general political representation. Thus different sections of the bourgeoisie-led Partido Revolucionario Institucional (Institutional Revolutionary Party—PRI), which in 1946 replaced Cárdenas’s Partido de la Revolución Mexicana (Party of the Mexican Revolution—PMR), expressed different developments in the accumulation process throughout the import-substitution-industrialization period. The ideologically diverse and pragmatic party-state included representatives of all segments of Mexican society under party-bureaucracy control (Maddison, 1993: 201–209; Collier and Collier, 2002 [1991]: 198–250; Knight, 2018).

Thus the policy and to some extent the ideological orientations of Mexico’s governments during the import-substitution-industrialization period largely mimicked those in Brazil and Argentina, showing however milder characteristics in all cases except during Cárdenas’s agrarianista endeavors. Thus, Cárdenas’s foundational nationalistic-populist period, which paralleled Vargas’s precorporatist (1934–1937) moves for workforce upskilling, was followed by a conservative shift that moderated those advances while solving infrastructure bottlenecks related to the post-Depression recovery as in Argentina and Brazil. The end of World War II brought into government, as in those countries, a nationalistic, anticommunist leadership that further increased state expenditures on infrastructure while enhancing state control of the labor movement. The 1950s saw the return of a revolutionary leadership when primary-commodity prices boomed and a movement from nationalistic to developmental populism when those prices dropped and foreign-capital inflows into manufacturing accelerated. This was followed by an authoritarian turn (1964–1970), as in Argentina and Brazil (Smith, 1990: 98–124; Maddison, 1993: 135–155; Knight, 2018: 256–257).

Despite these similarities, and as an expression of its mildness and the more stable flow of ground rent (Figure 2), the Mexican developmentalist phase (1955–1970) was distinguished by steady economic growth, low inflation, and low social conflict and was known as “stabilizing development.” Nevertheless, the 1970s period of increasing primary-commodity prices, expanding rent-bearing offshore oil production, and a related increase in foreign debt saw the consolidation of Latin American–style nationalistic populism and an increase in state-led import-substitution industrialization as the political and economic forms of capital accumulation. Echeverria’s Shared Development Program (1970–1976) and López Portillo’s Global Development Plan (1976–1982) echoed those implemented during Brazil’s 1974–1979 abertura and Argentina’s 1973–1974 Peronist revival (Fitzgerald, 1985: 212–218; Maddison, 1993: 155–159; Knight, 2018: 255–259).

As in Argentina and Brazil, the early 1980s crisis manifested itself in the dismantling of state-led import-substitution industrialization and took the politico-ideological shape of neoliberalism (Maddison, 1993: 159–166). In contrast, however, the sharp real-wage decline that followed enhanced the potentiality of a new modality of capital accumulation based on the production of industrial goods for world and especially U.S. markets, using a cheap and disciplined labor force, especially female, for simple manual-assembly operations. The maquiladora industry began to expand sharply in the mid-1980s after a decade of sluggish growth. The role of the Mexican economy in the global production of surplus value increasingly revolved around this new base thereafter (Grinberg, 2010).

Colombia: From Liberal Republic to National Front

Contrary to growing neoliberal/new-institutional-economics consensus, Colombian industrialization did not depart from the Latin American norm of a populist alliance between the industrial working-class and domestic-markets-oriented manufacturing capital that allegedly increased wages and profits, favoring inefficient industries against comparative advantages in agrarian/mining production (Urrutia, 1991; Kalmanovitz and Lopez, 2006). Substantial resources were transferred through state actions, including exchange-rate overvaluation, from the agrarian to other sectors of social production (García-García and Montes Llamas, 1989; and see Figures 1 and 2). Though coffee-export taxes were partly administered by the Federación Nacional de Cafeteros (Federation of Coffee Growers), this did not make ground rent exempt from them, since taxes were proportional to output while the federation’s para-state subsidies and services were evenly distributed. Nor did the low-interest credit and market protection afforded to agrarian production after the late 1950s mean subsidization; the former was necessary to induce inputs-intensive cost reductions and increase the ground rent available for appropriation, while the latter only partly compensated for the impact of exchange-rate overvaluation and state regulations on domestic prices and profits.

Rather, the singularities of the Colombian political economy sprang from the distinctive conditions of ground-rent recovery through import-substitution industrialization. Thus, with less ground rent (because of less favorable soil/weather conditions and the higher costs involved in transporting primary commodities through rugged terrain), import-substitution industrialization was quantitatively less developed than in the other large Latin American economies (see Figures 2 and 3). Moreover, as in Mexico (though starting a decade later and developing to a lesser degree), capital accumulation emerged through relatively extensive state-mediated, U.S.-funded development of intensive agriculture to increase the supply of foodstuffs and, crucially, agrarian-origin industrial inputs (García-García and Montes Llamas, 1989: 36–58; Kalmanovitz and Lopez, 2006: 184–195) at below-international prices and thus expand the ground rent appropriated by capital invested in the domestic economy.

Thus, as an expression of its limited potentiality, the genesis of Colombia’s import-substitution industrialization and its subsequent reproduction came about through less extensive state regulation of economic processes than in Argentina, Brazil, and Mexico. As in those countries, though somewhat later, state-owned enterprises were created in public utilities and oil extraction/refining, but, as in Mexico, relatively fewer were created in the heavy and engineering industry. Subsidized loans were granted by state institutions to domestic-markets-oriented industrial capital, although in smaller quantities. The tax system was also used to transfer ground rent to industrial capital through market protection and fiscal credit, although these surpluses were fewer than elsewhere (Gonzalez, 1990: 54–67; Ocampo and Tovar, 2000: 245–263). Thus the limited coherence and assertiveness of state actions supporting industrialization did not, as is often argued (Urrutia, 1991; Kalmanovitz and Lopez, 2006), result from the ideological views of Colombian public officials or the greater bargaining power of agrarian landowning capitalists. Rather, they expressed the limited need of global capital for such nation-state policies for appropriating ground rent in the Colombian economy.

The limited potentiality of rent-sustained, state-led import-substitution industrialization manifested itself in limited working-class political power. Not only did industrial-sector capital need a workforce with simpler productive attributes and therefore less consumption power (see Figure 4) but the small relative size of that workforce further reduced the need for the national social capital to expand, through state mediation, working-class consumption of collective goods to take advantage of economies of scale in production. Hence the prevalence of safety-net clientelism instead of more extensive populist welfarism (Urrutia, 1991). Moreover, the comparatively low level of direct state regulation of social reproduction, mediating the transformation of ground rent into profits for industrial capital, meant that working-class agency did not play the same active role in pushing for national development as it did in 1940s–1960s Argentina, Brazil, and, to a lesser extent, Mexico. In short, the weaker/milder populist movement/alliance did not express a differently structured process of capital accumulation or a distinctive cultural/institutional trait of Colombian society. Rather, it expressed the limited need for this process to come about through labor-led political actions lobbying the state for direct regulation of working-class reproduction, market protection, and the centralization of capital.

Thus Colombian political singularities were a matter of degree/quantity rather than substance/quality. In parallel to Cárdenas’s and Vargas’s governments in Brazil and Mexico, extensive labor protection laws and welfare-enhancing reforms were introduced by the 1930s Liberal administrations, especially Alfonso López Pumarejo’s Revolution on the March. These transformations also came about through state-directed strengthening of unions and an upsurge in their struggles for those advances (Abel and Palacios, 1991a: 596–598; Collier and Collier, 2002: 289–295). As in the other countries, these political processes mediated the formation of an incipient industrial working class with the skills, however limited, required for capital accumulation through import-substitution industrialization. As in contemporary Brazil and Mexico, the moderate Liberal government (1938–1942) that followed the populist experience moved to foster capital formation in the basic and primary-transformation industries. To that end, it created specialized state institutions that used a portion of the ground rent captured through taxes on foreign trade to fund private- and public-sector investments in manufacturing and enhanced the protective system that, combined with multiple exchange rates, channeled another portion of the ground rent to industrial capital producing nondurable consumer goods for domestic markets (Gonzalez, 1990: 56; Ocampo and Tovar, 2000: 245–247).

Once those bases were consolidated, Liberal governments gave way, as in Brazil and Mexico, to Conservative administrations that limited wage growth. Moreover, despite its distinctive characteristics, during the Korean War commodities boom, when the inflowing ground rent sharply expanded, the politico-economic forms of capital accumulation closely resembled Argentina’s and Brazil’s nationalistic-populist developments (Abel and Palacios, 1991b: 624–626). After a failed attempt to gain power by the Liberals’ left wing that ended in the killing of their leader, initiating a ten-year civil war in rural areas, the de facto Gustavo Rojas Pinilla government carried out most of their program, undertaking large investments in education and infrastructure while increasing state expenditures in social wages. As in Argentina and Brazil, the process came about through the short-lived consolidation of a charismatic military-strongman leadership and state-controlled-union advances. Similarly, it ended with the end of the boom and the ensuing decline in inflowing ground rent. As this happened, foreign industrial capital, especially of U.S. origin, accelerated its entry into the Colombian economy to take advantage of the conditions created under the nationalistic-populist governments, though in smaller numbers and with less complex production than elsewhere (Gonzalez, 1990: 56–58; Abel and Palacios, 1991a: 606–608).

The singularities of the Colombian process of capitalist development were also apparent in particularly high levels of noninstitutionalized social conflict. Not only did the inability of industrialization to absorb the growing rural workforce, together with rapid technological change in agrarian production, result in the consolidation of a large surplus population in rural areas (Ocampo and Tovar, 2000: 267–271) but, once the frontier lands were opened through state-managed colonization programs, the centralization of capital to take advantage of economies of scale and scope came about through the politically mediated elimination or displacement of peasant families (Machado, 1988: 258–274). In contrast to the situation in Brazil, most of these were semi-independent petty-commodity producers rather than plantation populations accustomed to the advances of capitalists and the state. In Colombia there was constant rural violence; initially along mainstream-party lines and subsequently involving more radical political options and the production and trade of rent-bearing narcotic substances (Kalmanovitz and Lopez, 2006: 365–374).

Once the first wave of postwar rural violence (1948–1958) had killed around 250,000 people and displaced many others, institutional stability returned by means of a power-sharing agreement between the two traditional political organizations, the National Front pact (1958–1974), and an open-ended state-of-siege. The mild character of Colombian import-substitution industrialization, resulting from the limited ground rent available for appropriation, meant that the process did not need to involve, as it did in Argentina, Brazil, and Mexico, new political parties and structural-break political processes, except for the short-lived high-rent periods of the mid-1950s and early-1970s commodities-price booms. Nor did working-class political participation need to be institutionalized through the party system. This alliance between the moderate sections of the Liberal and Conservative parties, with their informally associated labor organizations representing different working-class sectors, became the political expression of the import-substitution-industrialization process during its developmental-populist stage (Gonzalez, 1990: 58–61; Karl, 2018). The mildness of this process also meant that the valorization of capital was relatively less rent-dependent and therefore policy-mediated economic swings were comparatively softer in Colombia, and so were the institutionalized political changes through which they came about, even as violent social conflict in rural areas built up throughout the period (Collier and Collier, 2002 [1991]: 289–313).

Through the mid-1970s, Colombian state-led import-substitution industrialization also entered crisis, but the Colombian experience tended to be milder than elsewhere in Latin America. This was not only because import-substitution industrialization had developed relatively less intensively there but also because, when international coffee prices began to fall in the late 1970s, increasing oil, coal, and cocaine production limited the contraction of the ground rent available for appropriation by different social subjects in the Colombian economy (Gonzalez, 1990: 70–82). Notwithstanding the revisionists’ claims about Colombian democratic stability, the crisis of import-substitution industrialization and the subsequent transformation of ground-rent recovery by capital into a pattern that resembled pre-import-substitution-industrialization modalities came about through a new wave of violent armed conflict, largely limited to rural areas, that produced a further 250,000 deaths.