The crash of 2008 has been claimed, by right and left alike, to signal the end of the ‘neoliberal’ era. This may, then, be an appropriate moment to take a retrospective look, through the lens of comparative political economy, at one of the principal planks of its programme over the past three decades: the process of privatization. As with the liberalization of finance, the initiative came from the United States: the American airline industry was deregulated under the Carter Administration in 1978. In the uk, Thatcher launched a massive wave of state divestiture, in an economy where public enterprises had been a basic feature of the postwar period; a similar process followed in New Zealand. During the 1980s, however, despite the free-market orientation of many governments across the world, privatization remained a localized policy, restricted principally to Anglo-American economies. Yet by the early 1990s it had begun to be implemented on a global scale—to the extent that in 1993, the Economist could crow: ‘a policy that in 1980 seemed adventurous to some and unworkable to everybody else is now economic orthodoxy worldwide.’footnote1

By the end of 2002, a total of $1.1 trillion worth of state assets had been sold, with the bulk of revenues going to industrialized economies; Europe led the way in terms of total value, though a larger number of transactions were completed in Asia.footnote2 But in the developing world, it was Latin America that sold more public assets by value than any other region, accounting for nearly 40 per cent of total proceeds outside the advanced capitalist states (see Table 1). Not only massive in scale, the process was conducted with staggering speed: while the uk sold 20 firms in the space of 10 years, for example, Mexico sold 150 in the space of six. With the exception of Chile, where large-scale state divestiture took place in the 1970s under Pinochet, privatization in Latin America was a phenomenon of the 1990s. Ownership and control of banks, telecommunications, oil, gas, petrochemicals and utilities such as water, transport and electricity, were sold as part of a privatization stampede that took in Argentina, Brazil, Mexico, Peru, Bolivia, Venezuela and Paraguay.

Article figure NLR55medeirospicture1

What explains the character of the privatization process—its speed and extent—in Latin America? In what follows, I will argue that the mass privatization of the 1990s was not conceived or undertaken as a pragmatic reorganization of state and market structures in response to genuine macroeconomic problems; rather, it was the result of a political-ideological decision to expel the state from business, irrespective of the sectors and markets involved, or of the need for a provider of public goods. External pressures certainly had a significant influence, as deeply indebted states enacted the structural reforms demanded by the ‘Washington Consensus’ in order to obtain new loans or the cancellation of debts. But it was first and foremost a domestic political decision, driven by new ‘distributional coalitions’ that emerged from the balance of payments crisis in the 1980s, and who then sought to rebuild the state in their own interests.footnote3 For the blanket privatizations that took place in Latin America have profoundly altered the core of the state, increasing the power of major domestic and foreign capital over the commanding heights of the economy.

Before mapping the process of privatization in Latin America and describing the formation of the new coalitions that drove it, it is first necessary to provide some assessment of the role of state-owned enterprises in economic development, both in Latin America and elsewhere, and the arguments offered in favour of privatization. State-owned enterprises have tended historically to be concentrated in infrastructure and basic industries, since the externalities involved affect the whole economic system: the state assumed the burden of providing a necessary input or capital good for a multipurpose use that exceeded its social cost in terms of subsidies. In many countries soes were established to promote technological advance; in Western Europe, for example, state infrastructure enterprises were created in the latter part of the nineteenth century as a strategy for reducing capital costs in the face of competition from Britain.footnote4 soes were particularly concentrated in natural resources—petroleum, gas and minerals—due to the large scale of operation, high risk, and a desire to ensure that the large rents obtained would accrue to the state. After the Second World War, state enterprises began to have a stronger presence in infrastructure—transport, electricity and telecommunications—along with steel, as a heavy industry with forward linkages, in Western Europe, Japan, East Asia as well as in less developed countries.

In Latin America, the development of state-owned enterprise took a variety of forms. In Mexico, Lázaro Cárdenas nationalized oil, electricity and the railways in the 1930s as part of his policy to promote economic development; a similar consideration was behind the nationalization of steel and oil in Brazil by Getúlio Vargas in 1941 and 1953 respectively. The nationalizations that took place in Mexico in the 1960s and 70s, meanwhile, were largely designed to rescue private enterprises in oligopolistic sectors from bankruptcy. In Brazil, on the other hand, the second wave of soes created in the late 1960s and 1970s—the aircraft manufacturer Embraer in 1969, the telecom monopoly Telebrás in 1972, Nuclebrás in nuclear energy in 1975—were part of a strong effort to promote heavy industry. The reasons for the nationalizations in Chile under Allende were more political and ideological—aimed at breaking the hold of the traditional oligarchy on the country’s economy and society. But across Latin America as a whole, nationalization of public utilities, telecommunications, oil and mineral production was a common policy, pursued from the 1950s to the 1970s as a means of fostering industrialization and promoting the national private sector in the face of foreign competition. In Brazil, Argentina and Mexico this policy attracted foreign capital interested in exploiting a growing internal market. In several countries, the military played a key role in setting up soes, and national security concerns were an important ideological and political argument for nationalization. But with the exception of military industry in Brazil and Argentina, the involvement of the army did not affect the industrial composition of state intervention.

The developmental states that superintended industrialization in Latin America concentrated their investment efforts in heavy industry and infrastructure. This forms a significant contrast with the most commonly cited examples of successful postwar development—South Korea, Taiwan and Singapore—where actual state-owned enterprise, though significant, remained subordinate relative to private-sector industry. However, state participation in industrial investment was actually higher in these countries than in latecomers such as Brazil or Mexico, due to the dominant role played by state planning and pilot agencies—a fact that is commonly understated in contemporary institutionalist interpretations of economic development.

What, then, was the relative economic weight and role of state-owned enterprises, worldwide, before the launch of mass privatization? By the 1970s, public enterprise accounted for an average of 13.5 per cent of capital formation in countries for which investment figures were available, according to a broad statistical comparison carried out by the imf.footnote5 The comparison did not include the us, where the figure exceeded 16.5 per cent; nor of course the Soviet Union or China. Otherwise, among developed countries the greatest proportions were observed in Australia, Austria, Canada, Italy, France, Norway and the Netherlands, countries that had high economic growth, as well as in the uk, with a slower growth rate. In Germany and Japan the figures were lower, but never below 10 per cent. Average output and investment shares were very similar for developing countries in Asia, South America and Europe.

Article figure NLR55medeirospicture2