Is China’s economic expansion responsible for pushing Latin America back in the periphery of the world economy?

Jean-Christophe Defraigne and David Villalobos

Introduction

This chapter will analyse to what extent China’s economic penetration in Latin America has transformed the role of Latin American economies in the global division of labour. The dramatic increase of Chinese trade with Latin America has fostered debates over the phenomenon of a relative deindustrialization labelled reprimarizacion. Indeed, China-Latin American trade patterns constitute a classic case of North-South relations. Latin America would find itself in the economic periphexy vis-a-vis China that would replace the traditional Western industrialized economies in the position of economic centre. Focusing on the large economies of Latin America (Argentine, Brazil, Chile, Colombia and Mexico) that developed a significant industrial base and large manufacturing firms in the 20th century, this chapter will show that the role of China is often overestimated and that reprimarizacion was already well engaged before the Chinese commercial penetration of Latin America. ^[1]

generate technological innovation capacities.² Production costs were extremely high as machinery costed the double or the triple that the amount paid by firms in the United States.³ Some of these production units were subsidiaries of the United States and Western European multinational enterprises (MNEs). These were satisfied with national protected markets in Latin America. This enabled them to give a second life to obsolete machines that could not provide a sufficient level of competitiveness in the more open international markets. Naturally, these MNEs were not interested in transferring the latest high-tech production methods to developing countries as the risk of expropriation continued to exist throughout the 1960s and 1970s.⁴

The debt crisis and the scaling down of ISI

For these reasons ISI policies generated increasing trade deficits as the industrialization did not generate new source of exports but required an increasing amount of technology and machine tools imports when Latin American economies began to develop capital-intensive manufacmring activities and services. Unlike its counterparts in Japan, South Korea or Taiwan, Latin American governments did not engage in serious industrial and R&D policy to enable local firms to upgrade their innovation capacities and to move up the value chain by creating their own international brands.

The brutal dwindling of international credit after the Volcker shock from 1979 to 1981 and the plummeting of commodity prices after the 1982 recession highlighted the unsustainability of ISI policies.⁵ Most Latin American countries turned to the IMF for financial assistance and had to adopt structural adjustment policies in the 1980s and 1990s. IMF and World Bank imposed a series of policies that came to be known as the “Washington Consensus” that reduced the role of state in the economy, engaged in massive privatization schemes, lowered tariffs barriers and reduced barriers to FDI.⁶

Most of the other Latin American economies opened up considerably to other MNEs that took over entire industries in the local economy. In the 1980s and 1990s, these governments accepted their traditional role in the global division of labour as provider of commodities on international market. They also considered that global MNEs from the most advanced economies had gained more intangible access than their national champions and did not believe that an industrial policy could reverse this trend. These governments accepted that MNEs would provide, through trade or through their Latin American subsidiaries, the bulk of goods and services in capital-intensive industries.⁷ They all embraced the privatization of major state-owned national champions in utilities, transport and energy. Argentina, Mexico and Chile went further in the opening to flows of capital, goods and services than most of the developing economies around the world. Mexico’s president Salinas suggested to join NAFTA in 1989, drastically increasing the exposure of Mexican industries to US competitors.⁸ Argentina’s president Menem dollarized the Argentinian economy, which strongly reduced the capacity of the state to pursue ambitious industrial policies, should another government questions

Menem’s path of economic development. Only the Brazilian government was sufficiently influenced by the lobbies of capital-intensive industries to continue to protect its national champions and the Brazilian subsidiaries MNEs from international competition in the 1980s and 1990s.⁹ Nevertheless, even in Brazil, the lobbying weight of the agro-business continued to strengthen its dominant position in the shaping of national economic policies throughout the 1990s and 2000s. This implied that government resources allocated to industrial policies were limited. Rather than investing massively in infrastructure to lower the cost of capital- intensive manufacturing production that are large users of energy, water and modern transport infrastructure, the successive Brazilian governments insured some level of protection for its national champions through FDI negative lists, technical barriers to trade or privileged access to procurements. Unlike their US, East Asian and North Western European counterparts, the Brazilian government only allocated a limited amount of spending in R&D programmes at a time when information and communications technologies, new materials, nanotechnologies and robotics were deeply transforming capital-intensive activities. Brazilian authorities, industrialists and academics engaged in the 1980s on a debate on the necessity for an industrial policy in informatics. This generated the protectionist law of 1984 for informatics products, but this failed to deliver significant results in generating competitive products or services that could compete on international market.¹⁰ Eventually, this infant industry could not survive the reduction of state intervention under the Collor de Mello administration.¹¹

The mass privatization programmes pursued by various governments in Latin America had mixed effects. MNEs, mostly from the United States and the European Union, had a far more advanced technology and management know-how than the inefficient former state-owned firms riddled by cronyism and lack of high-tech investment. MNEs could deliver better sendees but often the privatization schemes replaced a state-owned monopoly by a private one that faced only weak national regulators and competition authorities. This situation gave MNEs a dominant position that enabled them to charge a high mark-up on utilities, telecom, transport and banking, reducing the level of international competitiveness of Latin American firms.¹²

The debt crisis and the subsequent structural adjustment policies of the 1980s and 1990s took their toll on human capital and infrastructure equipment. Allocations to education were reduced and the level of education stagnated during these decades at the time of rapid technological changes.¹³ While international repoxts highlighted the progress of education of enrolled students in sciences and maths in East Asia or in some emerging economies such as India, Turkey or Israel, they also revealed the relatively poor performance of many Latin American economies. These two decades were also characterized by a low level of investment in infrastructure that created a significant gap between Latin American economies and comparable emerging economies in East Asia or in the periphery of the European Union (the Mediterranean and Central European member states that joined between 1981 and 2004). Latin America and North America are the worst-performing regions in terms of investment gap in infrastructure according to numerous studies.¹⁴

The suppression or scaling down of ISI policies as well as the lack of positive industrial policies has generated a receding share of inanufacturing in the national economy that was visible before 2000 (Figures 4.l^t.6). Figures on R&D spending (Figure 4.7), royalties from intellectual property (Figure 4.8) as well as the number of highly ranked universities (Figure 4.9) were already pointing out to

Figure 4.1 Argentina Source: World Bank (2019).

Figure 4.2 Brazil

Figure 4.3 Colombia

Figure 4.4 Chile

a technological gap with East Asia and Western economies. With relatively high wages and production costs for developing countries, the large economies of Latin America had no comparative advantages in labour-intensive manufacturing. With their limits in terms of innovation and technology, they could not challenge foreign firms in international markets in technology-intensive manufacturing ox- services.

The relative decline of the manufacturing sector in the economy was less sharp in Mexico (Figure 4.5), thanks to its insertion in the regionalized production

Figure 4.5 Mexico

Figure 4.6 Manufacturing value added as percentage of the GDP (1965-2017) Source: World Bank (2019).

networks of MNEs using the NAFTA framework, notably in automotive and electronics. Nevertheless, Mexican firms in manufacturing played the junior partner- role in the production networks controlled by US, European or East Asian MNEs. The largest Mexican companies are conglomerate in energy, utilities, retailing, banking and telecom that operate more nationally than internationally with the exception of Grupo Bimbo, a bakery product manufacturing with a presence on

Figure 4.7 R&D spending in percentage of GDP

Source: World Bank (2020), Malaysia figure for 1997 and 2007 are not available, the figures on this graph are, respectively, those for the years 1996 and 2008.

Figure 4.8 Royalties for Latin American countries and other regions in the world (2005-2013)

Source: World Bank (2020).

Figure 4.9 Number of highly ranked Latin American Universities Source: Tunes Higher Education, University Ranking (2020).

various continents but clearly not characterized as high-tech activities. Gallagher and Porzecanski have synthesized this situation:

Built-up during the ISI, Mexican electronic firms were virtually eliminated after trade liberahsation and replaced by a foreign enclave economy with few linkages, minimal R&D (.. .) The domestic high-tech industry is nearly extinct (...). The assessment characterises Mexico as having a ‘maquila innovation system’. This is a system that imports technology and equipment and hosts networking activities by MNCs (MNEs) in a manner divorced from the broader economy.¹⁵

South American economies have therefore been on the path of relative technological decline, of deindustrialization and reprimarizacion since the 1990s, that is, before China acceded the World Trade Organization (WTO) and became a significant trading partner for Latin America (Figure 4.6).

Manufacturing value added as % of the GDP (1965-2017)¹⁶

The Chinese commercial presence in Latin America in the trade for goods

The objective of this section is to assess the evolution of the Chinese commercial penetration of these largest Latin American economies during the last five decades

Figure 4.10 High-technology exports (% of manufactured exports, 2000-2015) Source: World Bank (2019).

Figure 4.11 Argentina’s imports contribution by origin (1962-2016)

(1962-2016). Drawing from the bilateral trade statistics,¹⁷ a simple descriptive analysis of the import structure of Argentina, Brazil, Colombia, Chile and Mexico identifies clear-cut trends. First, China’s contribution to Latin American imports became significant only in the 2000s, after China’s accession to the WTO in 2001 (see Figures 4.11-4.15 and Table 4.1). After 15 years, China accounts for 19.47%

Figure 4.12 Brazil’s imports contribution by origin (1962-2016)

Figure 4.13 Chile’s imports contribution by origin (1962-2016)

ofArgentina’s imports, 17.93% ofBrazil’s, 25.82% of Chile’s, 21.45% of Colombia’s and 18.74% of Mexico’s. Second, the US, Europe and even other Latin American countries constimte the largest source of imports than China. Even if their contribution has been declining over the last two decades, it is still larger than China.

Figure 4.14 Colombia’s imports contribution by origin (1962-2016)

Figure 4.15 Mexico’s imports contribution by origin (1962-2016)

Table 4.1 Average imports contribution by origin

Source: Authors’ calculation from data from the Observatory of Economic Complexity (2019).

The Chinese commercial presence in Latin America in terms of FDI and sendees

Trade is not merely the trade balance for goods. Services account for a greater part of the economy around the world and in Latin America. Firms from developing countries usually opt for exporting their final product as they possess limited international management know-how and few or no intangible assets.¹⁸ As manufacturing firms develop a larger scale, a greater experience in production methods and learning-by-doing effects, they might opt for FDI to penetrate overseas market rather than simply export the final good. The MNEs from the economically most advanced nations will create subsidiaries in a given region that will organize their production network using many local inputs. Since the 1990s, companies like Danone, VW or Renault have regionalized their production process in different regions like the European Union, NAFTA or Mercosur, relying largely on local inputs and taking advantage of existing regional agreements.¹⁹ These transformations generate relatively less trade between the most developed economies and Latin America but more trade within Latin America. European and US MNEs have been present in manufacturing activities for decades in Latin America, sometimes as early as the launching of ISI programmes.²⁰ These incumbent firms are already producing most of their components locally for the Latin American markets, contrarily to the bulk of Chinese firms, which mostly produce in China and then export their finished products. Furthermore, one should not forget that some of Latin America’s manufacturing imports from China are not made by Chinese firms but by non-Chinese MNEs operating in China. The most common estimates of the share of China’s exports made by foreign-based MNEs put it at around 40%.²¹

The same remarks apply to services and utilities. The privatization waves of the 1980s and 1980s induced by the IMF structural adjustment programmes, generated numerous take-over of former state-owned national champions by US and EU MNEs.²² Often, this presence of US and EU MNEs will not be highly visible in trade statistics. However, these MNEs enjoy now an incumbent position in industries characterized by high barriers to entry or natural monopoly position, making it extremely difficult for local enterprises with limited experience and funding to enter the market without state support. In banking, protection has allowed local large firms, notably in Brazil, to resist foreign competitors. Chinese state-owned banks play a decisive role in the financing of Chinese subsidiaries operating in Latin America as well as large infrastructure projects carried out by Chinese firms.²³ Nevertheless, their rise has slowed down since 2017 and their role in terms of providing financial services in the region is far more limited than the European and US incumbents such as Banco Santander, BBVA, Citibank or BNP-Paribas.²⁴ In utilities and telecoms, Chinese firms have managed to penetrate Latin America through M&A and joint ventures.²⁵ Distribution also shows a dominant position of Western incumbents such as Carrefour, Walmart or Casino.

These phenomena explain why an analysis based solely on the trade of goods, neglecting services and FDI is likely to overestimate the economic influence of emerging economies, most notably of China and underestimate that of incumbent MNEs from Europe and the United States that have been present for decades. If one wishes to measure the impact of Chinese trade on Latin America’s reprima- rizacion, it has to look at all forms of commercial penetration on Latin American markets in manufacturing and capital-intensive industries (like telecom, energy and utilities). Latin American firms are not pushed out of secondary and high-tech tertiary sectors by Chinese firms alone. As Figure 4.16 indicates, FDI flows in Latin America reveals the dominant position of US and EU MNEs compared to Chinese ones.

Figure 4.16 FDI - net incurrence of liabilities inflows - equity capital by origin a. Brazil

Source: Banco Central do Brasil (2019).

Figure 4.16 b. Argentina

Source: Banco Central de la Repiiblica de Argentuia (2019).

Figure 4.16 c. Colombia

Source: Banco de la Repiiblica de Colombia (2019).

Figure 4.16 d. Chile

Source: Banco Central de Chile (2019).

Figure 4.16 e. Mexico Source: Gobiemo de Mexico (2019).

2 The latifundia comprador bourgeoisie and the absence

of coherent R&D and innovation policies

By the late 2000s, Chinese commercial penetration in Latin America was visible. The phenomenon of reprimarizacion continued, albeit at a slower pace than in the 1990s. The collapse of commodity prices in 2011 highlighted once again the dependency of Latin American economies on their primary exports. As the Brazilian and Argentinian economies receded, the debate over reprimarizacion gained momentum with a strong focus on the rise of Chinese exports.²⁶ As the previous sections of this chapter have shown, the role of China in the relative deindustrialization that South American economies have been experiencing for four decades is limited.²⁷

However, Chinese exports might have hurt specific labour-intensive sectors across Latin America. While many US and European MNEs focus on the high value-added upper segment of the market, many Chinese firms (and firms from the rest of developing Asia) exploit their comparative advantages on labour-intensive products. By targeting the lower end of the market, Chinese firms must have hurt low-tech Latin American manufacturing producers more than European and US competitors that based their production on different comparative advantages. Toro Hardy argues that local producers were displaced by Chinese competitors in machinery, textile, garment, shoes, rubber, pharmaceutical and plastic products.²⁸ However, this effect only reinforced but did not generate the deindustrialization experienced by South America for four decades.

What is particularly significant is the absence of coherent reaction from South American governments since the downturn of 2011. First, Latin America cannot rely on cheap wages to develop a manufacmring sector because the general level of price and Dutch disease effects make local firms less competitive than those from developing Asian economies such as China, not to mention Vietnam or Bangladesh. At such level of domestic wages, input prices and economic development, South America has to improve productivity. This requires the upgrading of domestic technological capacities. No major and consistent programmes to upgrade R&D capacity or science curriculum were adopted across Latin America. This region remains with one of the least advanced in the world regarding R&D intensity, IP royalties and scientific publications.²⁹ Second, countering deindustrialization would also necessitate a much higher level of investment in infrastructure in transport, telecom and energy to reach a level comparable to East Asia.³⁰ Again, no significant progress has been achieved these last decades. The share of GDP spent in overall infrastructure plummeted from 5.4% in the 1970s to 2.1% in the 2000s.³¹ Toro Hardy outlines the role of structural adjustment programmes in the 1980s and 1990s to explain an infrastructure gap between Latin America and Asia and shows the incapacity of the region to close that gap in the 2000s.³² The private sector remains passive and the government is more constrained financially by the macroeconomic effects of the crisis. These elements point out the lack of positive horizontal industrial policy across the region.

In Brazil and Argentina, the governments that have been elected after the recession of 2014 have opted for less state intervention in the economy. Since his election in 2015, President Macri prioritized the restoration of Argentina’s creditworthiness on international financial markets and lifted tax on agricultural exports to please the agribusiness. It is too early to assess what will be the effective choices of the Bolsonaro government regarding industrial policy. As a candidate, the new Brazilian president expressed contradictory messages on this topic.³³ However, his choice to appoint Paul Guedes as minister for economics that includes budget, industry and commerce is significant. Guedes completed his PhD at the University of Chicago, taught at the Universidad de Chile during Pinochet dictatorship and is a proclaimed disciple of Milton Friedman.³⁴ He believes that the solution to restore growth in Brazil is to move towards “a market-driven economy instead of the failed dirigiste economy that corrupted the political order"}- In 2019, Bolsonaro administration has let Boeing take over 80% of Embraer, a major high-tech national champion of Brazil.³⁶ Chile has long given up the concept of industrial policy and opted for a laissez-faire in terms of economic policy, multiplying bilateral FTAs.

One can naturally observe the corruption and the patronage that has been characterizing the relations between politicians and national champions like Petro- bras, Odebrecht, PVDSA or PEMEX. This is not specific to Latin American economies. However, Latin America seems to be the region where numerous mainstream economists and right-wing politicians reject state interventionism and opt for a Chicago-style neoliberal economic doctrine based on endorsing the “Washington Consensus”, accepting the current comparative advantages, which means keeping Latin America as a provider of commodities in the global division of labour. Among most emerging economies, notably in Asia (China, India, South Korea, Indonesia, Turkey), the dominant economic currents are far more interventionist.³⁷ Even in some of the most advanced economies, the notion of industrial policies had gained momentum since the 2008 crisis, notably in the United States and also recently in Germany.³⁸ Free market supporters opposed to state dirigisme have always existed in most capitalist societies, but they are far more prominent in Latin America and receive the political and media support of a large segment of the business community. A long-term structural cause can explain this limited support for industrial policy across Latin American economies.³⁹

The bulk of the Latin American capitalists has been engaged into primary sector activities, agriculture and mineral resources these last two centuries. This group has always favoured a Ricardian approach to international trade and never- favoured the use of government funds to support an ambitious industrial policy. It was only in exceptional circumstances, the crisis of the 1930s and the Second World War that the traditional Ricardian trade policies could not function because of the fall in international prices. This situation made the latifundia oligarchies lose the political initiative.⁴⁰ The worsening economic and social situation and the risk of communist subversion⁴¹ made the latifundia oligarchies accept a more authoritarian (or state interventionist in the case of Cardenas in Mexico) form of government. The governments that emerged from the 1930s and 1940s in Brazil, Argentina or Mexico would impose most state interventionism without fundamentally confronting radically the latifundia lobby or the foreign-based MNEs from the United States and Western Europe.⁴² They also managed to do this because of the geopolitical context of the Second World War effect. For example, Vargas managed to convince the United States to finance and provide technology transfer to create Brazil's modem steel industry.⁴³

Soon after this exceptional economic and geopolitical crisis of the 1930s and 1940s, these regimes gave way to more traditional administrations that reverted in part their economic dirigisme. Most of the traditional business elites would not invest in domestic manufacturing projects which was mostly done by the state and through state procurement such as Embraer. In Brazil, many of the success in terms of industrial and technology upgrading policies were actually achieved by the state and the military, not the private sector.⁴⁴

The wealthiest Latin American families continued to focus on the primary sector and also began to develop in the low-tech services such as hotel, food and drinks, restaurants, distribution or services linked to state procurement ox- regulation such as public works, utilities, postal services, media. It does not mean that they have modernized their production methods. The Brazilian, Chilean, Mexican and Argentinian agribusiness have developed very competitive techniques in planting, harvesting and food processing. However, most do not dare to confront the US and European incumbent MNEs in the high-tech industries. Despite the suggestion from Brazilian experts and civil servants that coffee bean producers should move downstream to develop activities in roasting, processing, develop their own distribution networks and brands, the industry leaders remain very timid in their response.⁴⁵ It is therefore not surprising that the wealthiest Latin American families were never attracted by Friedrich List’s notion of infant industry protection or by the various theories that justify a developmentalist state. These ideas were more supported by the civil servants, the military or some of the labour unions, who seldom could impose them on the traditional comprador latifundia lobby.

Many scholars have outlined how a major sociological and economic longterm characteristic of the wealthiest Latin American families has been their financial and cultural connections with the US and Western European countries.⁴⁶ Many hold portfolios or funds that invest globally, notably on the Western European and US stock exchanges. Many of their children will benefit from an education made in highly ranked universities in the United States or Europe and some enjoy a second residence in these countries. As part of a cosmopolitan Western bourgeoisie, what Gunder Franck referred to as the "international bourgeoisie” of Latin America,⁴⁷ sometimes holding important financial interest in US or European MNEs, they do not feel attracted by economic nationalism and are naturally neoliberal. They are usually in favour of free capital flows and support policies that guarantee the possibility to invest their capital gains anywhere in the global financial system (notably, the convertibility of their currency, pegging it to the dollar or a currency board). They are alien to the notion of dirigisme, capital control and economic sovereignty (even though they might adopt a rhetoric that supports political sovereignty). They naturally did not oppose the IMF structural adjustments programmes and the "Washington Consensus” that locked in these economic goals.

Latin America is not exceptional in this regard. ISI and FDI managed to generate industrialization outside the Western economies but, in most developing countries, under the control of MNEs from the most technologically advanced economies (the United States, Western Europe and Japan).⁴⁸ Nowadays, only a handfitl of developing countries have national private firms that have become global MNEs, capable of controlling international production networks and that are innovators setting the new technological standard for their industry. Manufacturing and high- tech services are usually provided by subsidiaries of MNEs or by local firms acting as junior partner within international production networks orchestrated by global MNEs from the most technologically advanced economies.⁴⁹

East Asia is often pictured as the counter-example with some national companies transforming into global MNEs in the 1990s (in the 1970s for Japan) and advanced innovation capacities reflected in large royalties from intellectual property, scientific publication and some local universities reaching the top 100 in international rankings. To be specific, this is the case for just six economies: Japan, South Korea, Taiwan, China, Hong Kong and Singapore. The two last ones were city-states, who benefited from an exceptional geographic position and a tax-haven regime that could not be compared with a typical developing economy from Latin America and elsewhere.⁵⁰ The first four ones experienced specific historical conditions that were not sufficient but necessary for their quick industrial take-off.⁵¹

They benefited from the geopolitical context of the Cold War in East Asia. The will from the US governments to contain the expansion of pro-Soviet regimes in East Asia generated the massive military interventions in Korea (1950-1953) and in Vietnam (1954-1972). To stabilize the socio-political situation among its closest allies in the region, Japan, South Korea and Taiwan, the US administration imposed land reforms that forced the local big landlords to invest in other sectors of the economy.⁵² The particular context of the defeat of Japan in the Second World War made this reform fairly easy for Taiwan and Korea, as the wealthiest families were Japanese colonialists that were expelled back to Japan in 1945.⁵³ The United States also provided these countries with substantial financial and technological assistance, including in some instance free use of US patents.⁵⁴ The United States also gave their East Asian allies a preferential access to their domestic market enabling them to adopt export-oriented growth and reach economies of scale in manufacturing. The United States actively supported the development of some heavy industry as a mean to build a strong industrial defence base for their immediate military needs in the Korean and Vietnam wars and to face a possible confrontation with the PRC. This US support enabled the military govermnents of Taiwan and South Korea to impose strong industrial policy in the 1970s and 1980s. They created the necessary infrastructure and R&D centres to develop a domestic heavy manufacturing sector and pressured some reluctant local capital holders to enter into capital-intensive industries. Japan had been an industrial power prior to 1945 but really developed a strong industrial sector with the war in Asia in the 1930s and 1940s.⁵⁵ Major industrial groups like Nissan or Mitsubishi developed fast thanks to the shift to the war economy and began to operate across the territories conquered. Taiwan and Korean firms also developed more heavy industry when the Japanese empire turned to a war economy in the late 1930s. Samsung was specialized in fruit and vegetable until 1939 when the Japanese military planning gave it the technology to provide electrical inputs for the wax- economy.⁵⁶

China had a different experience but also benefited from a revolution and from the Cold War. Many of the traditional most influential landlords fled with the Kuo- mintang (KMT) leadership to Taiwan. The revolutionary situation in the countryside made it easy for the Communist Party of China (CPC) to eradicate the remaining landlords in a bloody land reform in the late 1940s and early 1950s. No landed oligarchies could oppose the major economic modernization pursued by the CPC leadership. The CPC skilfully used the Cold War confrontation to access advanced technology. First, in the 1950s it relied on the USSR to create a capital-intensive industiy that was almost inexistent (apart from Japanese production units in Manchuria).⁵⁷ Steel, electrical power and modern infrastructure were built thanks to massive technological transfer from the Soviet Union. Then, when the CPC opted for a rapprochement with the United States in 1971, it also benefited in the next two decades from technological transfers of the United States in order to upgrade military capacities against the Soviet Union.^5S

Vargas in Brazil also played on the geopolitical context of the Second World War to receive US support for the creation of a national steel industry. Nevertheless, this phenomenon remained exceptional in Latin America. Far away from USSR and the PRC, Latin America did not experience major Cold War-related conflicts as East Asia. The US administration worried about communist subversion since the late 1940s and it became an important source of concern for the US government after the Castrists seized power in Cuba at the end of 1958.⁵⁹ However, the focus was to organize the fight against subversion by means of repression of the local states against a part of their population that was not organized militarily. Building a strong industrial military base and transferring technology to help US operation was not necessary in Latin America. The US Army and the CIA focused on training tens of thousands of Latin American officers in the School of the Americas and on Kennedy's so-called Alliance for Progress but it generated no major effects in terms of industrial policy and technology upgrading.⁶⁰ Disregarding suggestions made by some US experts, the US government also never imposed a radical land reform in any Latin American country. Such a move was difficult as it would have certainly met more resistance than their experience in East Asia given the entrenched position of the latifundia oligarchies in the Latin American society.

So, unlike the successful East Asian economies that benefited from an exceptionally lucky combination of factors, the large Latin American countries did not enjoy a similar evolution in terms of industrialization. Their landed oligarchies continued to play the role of a comprador bourgeoisie that accepted the current international division of labour, refusing a strong dirigisme and eventually liberalizing trade and capital flows in the 1980s, closing the parenthesis of the ISI launched in the late 1930s and 1940s.

The period characterized by a centralization of state power and the development of an industrial policy in Brazil, Argentina and Mexico did not last more than two decades. Most governments gave a priority to the traditional trade and monetary policies that suited the agribusiness controlled by the traditional latifundia oligarchies. Even Vargas and Peron could not break away from this path for their second term of presidency. Governments maintained some state-owned national champions in the 1960s and 1970s, but they accepted the turn to neoliberalism in the 1980s after the eruption of the debt crisis. The governments of Cardoso, Menem and Salinas all reduced considerably the state support to national industry.⁶¹

The global economic context of the 2000s reinforced further the lobbies of the agribusiness and the extraction industries across Latin America, regardless of the political rhetoric of various governments. The fast growth of emerging economies from Asia, mostly from China, generated an additional demand for raw materials and food products. Combined with a strong demand in developed economies boosted by the real estate and financial bubble of 2001-2008, this generated a rise of international prices for commodities price.⁶² Agribusiness lobbies strengthened with their fast growth and then contribution to the trade balance. Reprimarizacion was gaining ground but it was not a source of major concern for governments across Latin America. The recession that hit South America after 2011⁶³ has generated a return of neoliberal govermnents with some political figures that fit well the sociological profile of the comprador bourgeoisie. The president of Chili, Pinera, holds a PhD from Harvard and has a real net worth of USS2.8 billion and held 90% of the TV channel Chilevision and 27% of LAN Airlines. Maori’s father owned the Socma Group, a conglomerate that prospered thanks notably to government contracts during the military dictatorship⁶⁴ and then took over the postal sendees during the privatization wave of the Menem administration. The Macri family conglomerate has also produced car under Peugeot, Fiat and now the Chinese Cherry. Paulo Guedes, Bolsonaro’s minister of the economy, is a founder of the private equity firm Bozano Investimentos.⁶⁵

Conclusion

The phenomenon of reprimarizacion and the peripheral position of South America in the international division of labour and in international production networks cannot be attributed to the Chinese commercial expansion. First, despite its fast growth since the 2000s, China is not the only important commercial power present in Latin America, especially if one considers FDI and services and the incumbent position of some US and EU MNEs in the region. Second, the phenomenon of reprimarizacion mostly took place in the 1980s and 1990s, a period during which China’s commercial presence was still negligible. The relative deindustrialization that Latin America has experienced since the 1980s can be explained by the limits of the ISI model but also by the unwillingness of the traditional Latin American economic elites to engage in high-tech industries and to challenge existing MNEs from the United States, Europe or East Asia. Their behaviour was that of a comprador bourgeoisie that stuck to the primary sector and accepted the traditional role of commodity provider that Latin America had in the international division of labour since the 19th century. The international context of the global capitalist economy, the geopolitical situation and the geographic position of the region combined with the behaviour of the Latin American comprador bourgeoisie generated a path dependency in terms of economic development. The analysis of the successful East Asian economies that broke away from this path highlights the decisive role of exogenous factors, notably at the geopolitical level. Latin America did not benefit from such factors and followed the path of the majority of the developing world economies. Latin American trade with China has only reinforced this existing path dependency. Should China have not exported its manufactured products to Latin America, other East Asian economies would have constituted a similar challenge to the less competitive Latin American manufaemrers.

The consequences of reprimarizacion and of Latin America’s path dependency are serious. First, in the United States, then in Europe and finally in East Asia, manufacturing activities generated many benefits for their domestic economies. Manufacturing activities are characterized by a faster growth of productivity than the primary or tertiary sector. These productivity gains have often been translated into wages that stimulated the domestic demands. The non-qualified labour force in Latin American economies cannot rely on the manufacturing industries to provide them with jobs. Instead, a large part has to turn to low-tech labour-intensive services that do not provide such growth in wages. In developing economies, these activities are often performed by an informal sector with highly precarious and lowly paid jobs. The shift to a high-tech sendees economy has so far been made by economies who had experienced a high level of industrialization and who possesses high-level universities. That is true of Britain, the United States but also Hong Kong, Singapore, Switzerland and Ireland.

Given these elements and considering the current capitalist international division of labour, Latin American economies risk continuing their journey on the path of a low-tech reprimarizacion with severe social consequences. The stagnating wages and precarity of the low-qualified jobs, the high level of inequality and the low level of public goods are likely to continue to produce the very high level of criminality that characterize Latin American societies, thereby generating a significant negative effect on growth.⁶⁶ Furthermore, given its limited diversification in terms of exports, the dependency of Latin America on international commodities prices and its Dutch disease effects should continue.⁶⁷ Combined with the speculative and volatile international financial flows, Latin American economies should experience again boom-and-bust cycles with socially painful adjustments, as those suffered by the working class and the other poorest social groups in the regular crisis that have hit the continent since the 1980s, notably since 2014. This is consistent with the long-term economic analyses that forecast a quasi-stagnation of the share of Latin America in the world economy for the next three decades (while that of Asia will more than double).⁶⁸ This is the cost that Latin America pays for being stuck in the role of global commodity provider.

Notes

• 1 Furtado and Berthe (1973:1031).
• 2 Idem, p. 1045.
• 3 Cortes Conde (2009:295).
• 4 Gunder Frank (1979); Tera (2016); Defraigne (2016).
• 5 Kolko (1987); Bulmer-Thornas (2003).
• 6 Stiglitz (2008).
• 7 Oman (1994).
• 8 Lustig (1998); Dawson (2006).
• 9 Vidal Luna (2014).
• 10 Tera and Canedo Pinheiro (2016).
• 11 Thery (2012).
• 12 Oman (1994); Defraigne (2016).
• 13 Toro Hardy (2013).
• 14 Ruiz-Nunez (2015); World Bank (2020).
• 15 Toro Hardy (2013).
• 16 Heads of States bottom bar in Figures 4.1-4.5 is a graphical estimation based on their actual terms of office. It does not necessarily include all the Head of States, only the most representatives from this article perspective.
• 17 Observatory of Economic Complexity (2019).
• 18 Dunning (1993).
• 19 Oman (1994).
• 20 Verhulst and Wintgens (2015).

• 21 Kroeber (2016).
• 22 Bulmer-Thomas (2003).
• 23 Myers and Gallagher (2018).
• 24 Khan (2017); Myers and Gallagher (2018).
• 25 Myers and Wise (2017); Ramos Becard (2014).
• 26 Myers and Wise (2017:252); Toro Hardy (2013:170).
• 27 The Mexican and some Central American economies such as Costa Rica do not experience this phenomenon due to their insertion in regional production networks developed in the NAFTA CAFTA region.
• 28 Toro Hardy (2013:170).
• 29 European Parliament (2017).
• 30 Defraigne (2016).
• 31 Roett (2016:143).
• 32 Toro Hardy (2013:212).
• 33 Leahy and Schipani (2018).
• 34 Mena (2018).
• 35 Rathbone and Schinpani (2019).
• 36 Espinoza (2019).
• 37 Studwell (2014); Chaponniere and Lautier (2015); Kurlantzick (2016).
• 38 Defraigne (2017).
• 39 Stiglitz (2008).
• 40 Skidmore and Smith (1997).
• 41 Especially after the political events of 1935 in Brazil.
• 42 Pang (2002:31).
• 43 Rouquie (2006:115).
• 44 Thery (2012:230).
• 45 MDIC (2014).
• 46 Harvey (2013:186); Galeano (2009).
• 47 Gunder Franck (1979).
• 48 Defraigne (2016).
• 49 Adda (2012).
• 50 Except maybe for Panama that does not perform as well.
• 51 Defraigne (2006).
• 52 Studwell (2014); Flat (2014).
• 53 Chaponniere and Lautier (2015).
• 54 Defraigne (2006).
• 55 Flat (2014).
• 56 Chaponniere and Lautier (2015).
• 57 Zhang (2010).
• 58 Leffler and Westad (2010).
• 59 Idem-, Julien (1968).
• 60 Skidmore and Smith (1997); Julien (1968).
• 61 Bulmer-Thomas (2003); Skidmore and Smith (1997).
• 62 Defraigne (2016).
• 63 Not Central America. Mexico and Central America have their business cycles that are more synchronized with the US economy, partly due to the regionalization of MNEs manufacturing production processes generated by NAFTA and CAFTA.
• 64 Verbintsky and Bohoslavsky (2016).
• 65 Biller etal. (2019).
• 66 Toro Hardy (2013:191)
• 67 Gaulard (2016:75).
• 68 Chateau et al. (2016).

References

Adda Jacques. 2012. La Mondialisation de I'economie, Paris, La Decouverte.

Banco Central de Chile. 2019. (https://si3.bcentral.cl/estadisticas/Principall/Estudios/SE/ BDP/ied.html).

Banco Central de la Republica de Argentina. 2019. (www.bcra.gov.ar/PublicacionesEsta disticas/Estadisticas_inversiones_directas.asp).

Banco Central do Brasil. 2019.(www.bcb.gov.br/ingles/econoinic/SerielhstFluxoInvDir_i.asp).

Banco de la Republica de Colombia. 2019. (www.banrep.gov.co/es/inversion-directa).

Biller David, Lucchesi Cristiane, Gamarski Rachel. 2019. “Brazil Super Minister Shoulders Weight of Bolsonaro Economy”, Bloomberg, March 15.

Bulmer-Thomas Victor. 2003. The Economic Histoiy of Latin America Since Independence, Cambridge, Cambridge University Press.

Chaponniere Jean-Raphael, Lautier Marc. 2015. Les economies emergentes d'Asie, Paris, Armand Colin.

Chateau Jean, Fontagne Lionel, Foure Jean, Johansson Asa, Olaberria Eduardo. 2016. Trade Patterns in the 2060 World Economy, Economics Department Working Papers No 1142, Commission Europeenne.

Cortes Conde Roberto. 2009. The Political Economy of Argentina in the Twentieth Centuiy, Cambridge, Cambridge University Press.

Dawson Alexander. 2006. First World Dreams: Mexico Since 19S9, London, Zed Books and Macmillan.

Defraigne Jean-Christophe. 2006.“Commercial Integration” in Report on East Asian Integration: Opportunities and Obstacles for Enhanced Economic Co-operation, Dieter Heribert (ed.), Paris, Notre Europe, pp. 56-92.

-. 2016. “La reconfiguration industrielle globale et la crise mondiale”, Outre-Teire,

revue europeenne de geopolitique, No 46, pp. 143-192.

-. 2017. Introduction a I'economie europeenne, Ouvertures economiques, Louvain-

la-Neuve, De Boeck.

Dunning John. 1993. Multinational Enterprises and the Global Economy, Wokingham, Addison Wesley.

European Parliament. 2017. The EU-Latin American Strategic Partnership: State of Play and Ways Forward, study by Gustavo G. Muller, Jan Wouters, Jean-Christophe Defraigne, Sebastian Santander, Kolja Raube, Brussels, Directorate-General for External Policies, Policy Department.

Espinoza Javier. 2019. “EU regulators Open Investigation into Boeing-Embraer Tie-Up”, Financial Times, October 4.

Flat David. 2014. The Japanese Economy, Third edition, Oxford, Oxford University Press.

Furtado Celso, Berthe Jean-Pierre. 1973. “Le monde latino-americain: les homines et les richesses ”, in Le monde depths 1945, Tome 2: Les pays pauvres et la naissance de nou- veaux mondes, Maurice Crouzet (ed.), Paris, PUF.

Galeano Eduardo. 2009. Open Veins of Latin America, London, Serpent’s tail, pp. 10-261.

Gaulard Mylene. 2016. “Les responsabilites de la desindustrialisation precoce dans la crise bresilienne actuelle”, Outre-Terre, revue europeenne de geopolitique, No 47, pp. 66-81.

Gobierno de Mexico. 2019. (https://datos.gob.mx/busca/dataset/informacion-estadistica- de-la-inversion-extranjera-directa/resource/50030f94-32cc-41e7-8100-81c632b58554).

Gunder Franck Andre. 1979. Capitalisme et sous-developpement en Amerique Latine, Paris, Maspero.

Harvey David. 2013. The New Imperialism, Oxford, Oxford University Press.

Julien Claude. 1968. L ‘Empire American!, Paris, Grasset.

Khan Eazaz. 2017, “Top 50 Latin American, Caribbean Banks by Assets”, S&P Global, April 20 (http://www.revistacobertura.com.br/site-2017/wp-content/uploads/2017/04/ MarketIntelligence_Top50banks.pdf).

IColko Joyce. 1987. Restructuring the World Economy, New York, Pantheon Books.

ICroeber Arthur. 2016. The Chinese Economy, Oxford, Oxford University Press.

Kurlantzick Joshua. 2016. State Capitalism: How the Return of the State is Transforming the World, Oxford, Oxford University Press.

Leahy Joe, Schipani Andres. 2018. “Bolsonomics: The Reform Plans of Brazil’s New President”, Financial Times, December 31.

Leffler Melvyn P, Westad Odd Arne (eds.). 2010. The Cambridge Histoiy of the Cold War, 3 vols., Cambridge, Cambridge University Press.

Lustig Nora Claudia. 1998. Mexico: The Remaking of an Economy, Washington, DC, Brookings Institution.

MDIC-Ministerio do Desenvolvimento, Industrie e Comercio Exterior. 2014. Reposiciona- mento estrategico das industrias processadoras de cafe. Brasil, Passo Fundo, Meritos editora.

Mena Fernanda. 2018. “Economista de Bolsonaro, Paulo Guedes viveu mudanca radical em Chicago”, Folha de S. Paulo, August 9.

Myers Margaret, Gallagher Kevin. 2018. Down but Not Out: Chinese Development Finance in LAC 2017, China Latin America Report, The Dialogue, March (www.thedialogue.org/ wpcontent/uploads/2018/03/Chinese-Finance-to-LAC-2017.pdf).

Myers Margaret, Wise Carol. 2017. The Political Economy of China-Latin American Relations in the New Millenium, London, Routledge.

Observatory of Economic Complexity. 2019 (https://atlas.media.mit.edu/en/).

Oman Charles. 1994. Globalization and Regionalization, Paris, OECD Development Centre.

Pang Eul-Soo. 2002. The International Political Economy of Transformation of Argentina, Brazil and Chile Since 1960, Basingstoke, Palgrave Me Millan.

Ramos Becard Danielly, Vieira de Macedo Bruno. 2014. “Chinese Multinational Corporations in Brazil: Strategies and Implications in Energy and Telecom Sectors”, Revista Brasileira de Politico Internacional, Vol. 57, No 1, pp. 143-161.

Ratlibone John Paul, Schipani Andres. 2019. “Bolsonaro’s Financial Guru Plans Free-Market Perestroika”, Financial Times, February 11.

Roett Riodan. 2016. Brazil What Eveiyone Needs to Know, Oxford, Oxford University Press.

Rouquie Alain. 2006. Le Brest I au XXIe siede: Naissance d’un nouveau grand, Paris, Fayard.

Ruiz-Nunez Fernanda, Wei Zichao. 2015. Infrastructure Investment Demands in Emerging Markets and Developing Economies, Washington, DC, World Bank Group.

Skidmore Thomas E., Smith Peter H. 1997. Modern Latin America, Oxford, Oxford University Press.

Stiglitz Joseph, Serra Narcis. 2008. The Washington Consensus Reconsidered, Oxford, Oxford University Press.

Studwell Joe. 2014. How Asia Works, London, Profile Books.

Tera Cristina, Mauricio Canedo Pinheiro. 2016. “La pohtique industrielle au Brasil”, Outre-Terre, revue europeenne de geopolitique, No 47, pp. 82-93.

Thery Herve. 2012. Le Bresil, Paris, Armand Colin, coll. Geographie.

Times Higher Education. 2020. World University Rankings, accessed on November 10, 2020.

Toro Hardy Alfredo. 2013. The World Upside Down: The Complex Partnership Between China and Latin America, Singapore, World Scientific Publishing Company.

Verbintsky Horacio, Bohoslavsky Juan Pablo (eds.). 2016. The Economic Accomplices to the Argentine Dictatorship, Cambridge, Cambridge University Press.

Verhulst Gilles, Wintgens Sophie. 2015. “Unbalanced Triangle” in China, the EU and the Developing World, Wouters Jan, Defraigne Jean-Chistophe and Burnay Mathieu (eds.), Cheltenham, Edward Elgar.

Vidal Luna Francisco, Klein Herbert. 2014. The Economic and Social Histoiy of Brazil Since 18S9, Cambridge, Cambridge University Press.

World Bank. 2020. World Bank Open Data, accessed on November 10, 2020.

Zhang Shu Guang. 2010. “The Sino-Soviet alliance and the Cold War in Asia, 1954-1962” in The Cambridge Histoiy of the Cold War, volume 1, Leffler Melvyn P. and Westad Odd Arne (eds.), Cambridge, Cambridge University Press, pp. 353-375.

• [1] The sequence of Latin America’s deindustrializationand China’s penetration in the region The industrialization of the large Latin American economies The industrial take-off in Latin America stalled with the 1930s global crisis. Thefalling commodity prices generated a severe trade deficit that forced the adoptionof import-substitution industrialization (ISI) in the large countries that had alreadydeveloped previously an industrial base and possessed sufficient financial capacities to subsidize an infant industry, mostly Brazil, Columbia, Chile, Mexico andArgentina.1 The Latin American ISI was unsustainable for two main reasons. First, unlikesome industrializing economies of East Asia, the Latin American governmentsdid not develop a strategy to increase the level of R&D and autonomous innovation capacities. Sheltered from international competition by high tariffs andnon-tariffs barriers and relying on the domestic market demand, the productionunits of the manufacturing and services industries had no specific incentive to