Can Powerful Dictators Escape ‘The Market as Prison’? The Case of Pension Privatisation in Pinochet's Chile

Abstract

The central role of economic elites in shaping public policy in Latin America has become increasingly clear. Yet most of the recent literature on the subject focuses on democratic contexts. This paper analyses pension privatisation in Chile as a case study for improving our understanding of business–state interaction in authoritarian contexts. Globally, the 1981 pension reform carried out during the Pinochet dictatorship became an example for pension privatisation elsewhere. Analysis of the policy-making process, based on novel empirical material, shows that from 1973 financial groups accumulated growing power which enabled them to first (a) defeat their opponents within the economic elite, (b) overpower their rivals within the state and, finally, (c) force Pinochet into passing pension privatisation legislation. Our results stress the need to include the study of different actors’ power resources – along with ideological issues and the regime structure – in attempts to understand the outcome of policy processes in authoritarian contexts.

El papel central de las élites económicas en la configuración de las políticas públicas en América Latina es cada vez más claro. Sin embargo, la mayor parte de la literatura reciente sobre el tema se centra en contextos democráticos. Este artículo analiza la privatización de las pensiones en Chile como un estudio de caso para ampliar nuestra comprensión de la interacción entre empresarios y Estado en contextos autoritarios. Globalmente, la reforma de las pensiones de 1981 llevada a cabo durante la dictadura de Pinochet se convirtió en un ejemplo de privatización de las pensiones en varios países del mundo. El análisis del proceso de formulación de esta política, basado en material empírico novedoso, muestra que desde 1973 los grupos financieros acumularon un poder creciente que les permitió primero (a) derrotar a sus oponentes al interior de la élite económica, (b) dominar a sus rivales dentro del Estado y, finalmente, (c) obligar a Pinochet a aprobar la privatización de las pensiones. Nuestros hallazgos enfatizan la necesidad de incluir el estudio de los recursos de poder de diferentes actores – junto con aspectos ideológicos y la estructura del régimen – para comprender el resultado de los procesos políticos en contextos autoritarios.

O papel central das elites econômicas na definição das políticas públicas na América Latina tornou-se cada vez mais claro. No entanto, a maior parte da literatura recente sobre o assunto centra-se em contextos democráticos. Este artigo analisa a privatização das pensões no Chile como um estudo de caso para melhorar a nossa compreensão da interação entre empresas e Estado em contextos autoritários. Globalmente, a reforma das pensões de 1981, levada a cabo durante a ditadura de Pinochet, tornou-se um exemplo de privatização das pensões em vários países do mundo. A análise do processo de elaboração desta política, com base em material empírico novo, mostra que desde 1973 os grupos financeiros acumularam um poder crescente que lhes permitiu primeiro (a) derrotar os seus oponentes dentro da elite econômica, (b) dominar os seus rivais dentro do Estado, e, finalmente, (c) forçar Pinochet a aprovar a privatização das pensões. Os nossos resultados sublinham a necessidade de incluir o estudo dos recursos de poder dos diferentes atores – juntamente com aspectos ideológicos e a estrutura do regime – para compreender o resultado dos processos políticos em contextos autoritários.

Introduction

‘Who will manage the money? That gets me nervous because I also know that several gentlemen are becoming millionaires in this country, and they have sent people [abroad] to study the [pension] system since they want to operate in it’, said General Augusto Pinochet in October 1980 during the opening of the key secret session in which the military Junta discussed – and ultimately passed – the proposed reform to privatise Chile's pension system.¹ Later during the same meeting, the powerful dictator was to reiterate his concerns several times: ‘I believe that a new pension system is necessary, but the fact that there will be so much money and [that] here we have a bunch of crooks [who might take advantage of the system] causes me anguish.’²

Pinochet thus revealed his significant misgivings regarding the radical pension reform which was to replace the pay-as-you-go system by a scheme of individual capitalisation.³ In the privacy of a secret session of the military Junta, the dictator, who ruled Chile with an iron fist for 17 years (1973–90), recognised the anguish caused by the prospect of a pension system being run by the private sector. Pinochet was fearful of the mammoth financial conglomerates (the ‘gentlemen’ and ‘crooks’) that controlled large swaths of the Chilean economy thanks to the policies implemented by his own government.⁴

How can we explain, then, Pinochet's approval of pension privatisation in spite of these fears and even against the recommendations of trusted advisors? Previous research has explained the adoption of the radical reform by reference to state autonomy and the concentration of power in the hands of Pinochet.⁵ This literature, nonetheless, has not addressed the central question of why Pinochet passed the reform in spite of his strong reservations and those of many collaborators. We argue that in order to make sense of this puzzle we have to pay more attention to the Chilean economic elite and its multifaceted sources of power. Using Charles Lindblom's metaphor,⁶ we contend that Pinochet became hostage to the market prison built with his own assistance through a series of market-oriented reforms. As we shall see, these reforms strengthened financial conglomerates vis-à-vis other segments of the economic elite and the state, enhancing their power.

Our aim is thus to analyse pension privatisation in Chile so as to improve our understanding of business–state relations in authoritarian contexts. We show that from 1973 financial groups accumulated growing power, which enabled them to (a) firstly defeat their opponents within the economic elite (i.e. traditional conglomerates), (b) overpower their rivals within the state (high-ranking corporatist officers), and, finally, (c) force Pinochet into passing pension privatisation. We study these outcomes by means of a comparative analysis across three stages of the policy-making process, namely: (i) agenda setting before the coup (1970–3); (ii) the design of a compromise bill including both neoliberal and corporatist ideas (1973–6); and (iii) the definitive implementation of pension privatisation (1976–81). Our study of the policy process relies on a diversity of sources, including the minutes of the military Junta's secret sessions and a hitherto unpublished secret document prepared by the regime's Budget Office (see note 80). We also draw on interviews with key players who participated in the policy-making process, including five former high-ranking government officials and ministers with strong links to the economic elite. Additionally, we undertook extensive archival research in the press records at the Biblioteca Nacional (National Library). We identified causal mechanisms and assembled the explanations of our argument by triangulating information across those multiple, independent data sources.

In rethinking the relationship between the economic elite and the Pinochet dictatorship in Chile, we make a broader contribution to the political economy literature on state–business relations in non-democratic environments. This contribution is particularly useful given both the importance of dictatorships in Latin America's past development and the current authoritarian backlash across the world. Most of the recent literature on business power focuses on democratic contexts,⁷ paying much less attention to the role of economic elites in dictatorships. Yet, as we will show, several sources of business power identified by authors working on democracies are also useful for the study of authoritarian regimes. By adapting the conceptual framework developed for studying business in democratic contexts to an authoritarian setting, we provide new insights into the channels and mechanisms through which economic elites might influence the policy process under authoritarian regimes. Our contribution is also empirical as we provide novel evidence – including unpublished secret documents from the dictatorship era – about a case of global interest. Chile represents the world's first and most notable case of pension privatisation and an inspiration to a number of Latin American, Asian and European countries.⁸

We proceed now with a review of the literature on business–state relations – paying particular attention to the Latin American context – showing its insufficient attention to mechanisms of influence exerted by businesses in authoritarian contexts. We then argue that the categories for the study of business power recently developed for democratic contexts can be adapted to non-democratic regimes. The bulk of the paper comprises an empirical analysis of the policy process (1970–81). It concludes with some reflections on the broader relevance of our research.

Business Power and the State under Authoritarian Regimes: An Incomplete Literature

During the 1970s, influential political economists studied the role of big business under authoritarian regimes. Much of this literature – covering topics from bureaucratic authoritarianism to dependency theory – followed a macro-historical approach in order to show how capitalist groups influenced political and economic development. These approaches were interested in explaining structural outcomes: in the case of bureaucratic authoritarianism, the challenge was to account for the emergence of bureaucratic dictatorships as the result of the process of late modernisation.⁹ Dependency scholars assessed the historical origins of different capitalist modes of production.¹⁰ These approaches often assumed business influence instead of providing supporting empirical evidence or illustrating the mechanisms by which such influence operated in practice. Similarly, other influential authors have developed typologies of authoritarian regimes, distinguishing them according to the role played by ideologies, modes of repression, and party and mass organisations, among other factors.¹¹ However, in these works the multiple mechanisms by which big business influences the policy-making process also remain understudied.

After the attention on business elites faded from Latin America's research agenda during the 1980s, the issue returned to the fore in the early 1990s.¹² The new literature paid more attention to specific policy processes and provided richer empirical accounts but failed to develop an adequate understanding of business–state relations under authoritarianism. This literature swung between theories of the ‘autonomous state’ (in which insulated policy-makers were able to adopt whichever policies they wanted)¹³ and others suggesting that big businesses shaped policy in military regimes.¹⁴

Similarly, much of the study of social policy in authoritarian contexts in Latin America adopted a state-centred approach even when considering the role of economic elites. For example, in her informative analyses of Chile and Uruguay, Rossana Castiglioni highlights the centralised structure of Chile's dictatorship and the ideological position of technocratic elites.¹⁵ According to this view, Pinochet's one-man rule facilitated the rise of the Chicago Boys to the upper echelons of the Chilean state.¹⁶ From such positions and with high degrees of autonomy, they were able to push for a radical re-structuring of the country's pension system. This literature thus presents social policy reforms as a battle over ideas and state power, paying much less attention to the role of business interests.

Eduardo Silva considers the role of business groups in the politics of policy change in Latin America more explicitly.¹⁷ Focusing on the study of market liberalisation during the Pinochet regime, Silva argues that the pace and content of neoliberal reforms was a result of changing relations between the state and different business coalitions. He characterises the process of economic restructuring according to changes occurring within both the state and the dominant business coalition that supported the dictatorship, which also influenced the patterns of interaction between the state and business. Although we draw on Silva's excellent research, we aim to provide a clearer understanding of the different means and mechanisms used by economic elites to advance their interests. Since he focuses on explaining large-scale policy shifts, we contend that the next step required to advance theory on business power in authoritarian regimes is the detailed, sustained analysis of specific policy-making processes.

During the last decade, there has been a surge in studies of economic elites in Latin America, following and extending path-breaking work in the Global North. The work of Tasha Fairfield has inspired a number of contributions that have studied the influence of big business in different realms such as redistribution and industrial and social policy.¹⁸ This scholarly work has focused precisely on the examination of the multiple means and mechanisms by which economic elites might influence governments’ decision-making. Rooted in Jacob Hacker and Paul Pierson's seminal work,¹⁹ Fairfield specifies a comprehensive set of mechanisms of business influence and guidelines for operationalisation, thereby providing analytical tools with stronger explanatory leverage for causal inference.

This operationalisation is mainly based on two categories for the study of business power: instrumental power and structural power (see Table 1). Instrumental power refers to deliberate, organised actions carried out in the political arena to influence the policy-making process.²⁰ Lobbying, funding political parties and editorialising in the press constitute some examples of collective actions carried out by economic elites using their sources of instrumental power. The second type is structural power, which arises from the myriad of investment decisions – based on profit-maximising criteria – taken by firms and capital owners. In a capitalist society, the state depends on businesses to invest and produce in ways that foster collective prosperity.²¹ Hence, governments may themselves halt reforms that can harm investors’ confidence, thus affecting investment and economic growth.²² In contrast to instrumental power, structural power does not require any capacity for collective action.

Table 1. Main Descriptive Aspects of Instrumental and Structural Power

Note: Underlining indicates those sources of power that do not translate from democratic to authoritarian contexts; bold indicates sources of power that are new to authoritarian contexts (i.e. do not exist in the democratic context).

Source: Authors’ elaboration based on Hacker and Pierson, ‘Business Power and Social Policy’; Fairfield, Private Wealth and Public Revenue in Latin America; Fairfield, ‘Structural Power in Comparative Political Economy’.

Fairfield applies the instrumental–structural framework to the study of tax politics in Latin America. She finds that taxing economic elites was difficult when either their instrumental or their structural power was strong, or when, in combination, those two types of power exerted a significant influence on policy-making. As such, along with her specification of multiple causal mechanisms, her contribution to the field lies in the recognition that instrumental power and structural power are mutually reinforcing. These two types of business power were traditionally seen as mutually exclusive in the literature.

However, Fairfield and the recent scholarly work inspired by the instrumental– structural framework have focused primarily on democratic contexts. These works yield insights into how economic elites exert influence in market democracies, with elections and political parties. They rely, for example, on the close links between economic elites and conservative parties. How useful are these concepts in non-democratic environments? How does business power operate in authoritarian settings, with outlawed political parties, shut-down legislative assemblies, and highly concentrated executive power?

The recent literature exploring these questions is scarce. To the best of our knowledge, there are no studies of business power in Latin America's military regimes that apply the instrumental–structural framework. We need to turn our attention to other regions to find studies that follow this approach. In particular, Jeffrey Winters analyses Indonesia's political economy during General Suharto's authoritarian regime (1965–98).²³ He assesses how business power influenced Indonesia's policy trajectory, accounting for variations in the state's responsiveness to those controlling private capital. Nonetheless, his work focuses on the analysis of structural power (with, besides, a strong emphasis on capital mobility), paying less attention to instrumental power.

Adapting the Instrumental–Structural Framework to Authoritarian Contexts

In this paper we adapt Fairfield's framework to understand how economic elites can achieve their policy preferences in authoritarian settings. Economic elites, in this context, wield their power to obtain the policies they prefer, which in turn are determined by their material interests. For instance, the financial sector tends to constitute an unambiguous proponent of pension privatisation because it allows it to become providers of retirement savings products.²⁴ In this view, then, economic elites pursue their preferences by influencing policy-making through different sources of power (e.g. recruitment of business representatives into government), which allow them to carry out a number of actions (e.g. direct participation in policy-making), which in the end result in observable implications (e.g. pension privatisation). The application of the instrumental–structural framework to authoritarian contexts requires the modification of some of the causal mechanisms shown in Table 1: we underline those sources of power that do not translate from democratic to authoritarian contexts, while sources of power that are new to authoritarian contexts are highlighted in bold. For the sake of brevity, the following discussion will focus on the analysis of those sources of power, actions and observable implications that change when we move the analysis from democratic to authoritarian contexts.²⁵

Given the nature of structural power, we do not expect substantial changes in the way this type of power might influence the policy process. As long as markets remain a key allocation mechanism, we expect structural power to exert a similar effect on democratic and authoritarian governments: in both cases policy-makers refrain from implementing reforms that could harm investment. Nonetheless, at this point we depart from Fairfield's operationalisation since we acknowledge the possibility that economic considerations might compel both authoritarian and democratic governments to implement policies in order to promote investment. As such, the pressure to restore or enhance business confidence might also act in the opposite direction to a credible disinvestment threat: it may compel governments to enact and implement reforms benefiting business interests.²⁶

The channel of transmission of structural power does differ between the two contexts. In democratic regimes, policy-makers chiefly avoid negative aggregate economic outcomes since they might be severely punished at the polls. Dictators are subject to the effects of economic performance but in different ways. For instance, a severe economic crisis could prompt the economic elite to shift their regime loyalty and promote a process of political change. Likewise, an economic downturn might activate popular sectors, with the potential of becoming a veritable threat to the regime's continuity.

In comparison with structural power, sources of instrumental power vary significantly in authoritarian contexts. In terms of relationships with policy-makers, partisan linkages and election to public office are sources of power that do not translate to authoritarian contexts. By contrast, in authoritarian regimes there are new avenues for business influence. In military regimes, members of the armed forces usually become policy-makers. This fact determines the need – and opportunity – for business representatives to curry favour with high-ranking officers. In this regard, we expect to find evidence of power-building investments carried out by members of the economic elite aimed at strengthening their relationships with the military (i.e. ties to the armed forces in Table 1).

As for resources, the main differences relate to media access. We expect that money, technical expertise and cohesion will be mobilised in similar ways in both authoritarian and democratic contexts. In the case of media access, however, the direction of change is not clear and will be highly context specific. As a norm, authoritarian regimes severely curtail the freedom of the press. The extent to which an authoritarian government controls and censors their domestic press will determine the effect on ‘media access’ as businesses’ sources of power. In cases in which authoritarian governments censor free press except for media outlets owned by the economic elite, we expect business power through media access be significantly augmented. Under these circumstances, conservative media would have substantial leeway to shape public opinion and determine the policy agenda.

In the next section, we will see how the interplay of these multiple sources of power determined the implementation of the 1981 pension reform in Chile.

Privatisation of Chile's Pension System: Analysis of the Policy Process

The following sections analyse the policy-making process that started in the early 1970s and led to the outcome we are trying to understand: implementation of the radical 1981 pension reform. The aim is to carry out a comparative analysis of three distinct stages, highlighting changes in the independent variables (business power, state structure and the interaction among them) that led to changes in our dependent variable (different stages of the neoliberal pension reform). In the analysis of each subperiod we will begin by describing the outcome to be explained and will then discuss changes in the independent variables.

First Stage (1970–September 1973): Pension Reform in the Making

The Outcome

During this period, the first outline of the ultimately approved pension reform was published. Months before the coup, top executives and neoliberal technocrats – selected and controlled by business representatives – had drafted an economic programme dubbed ‘El Ladrillo’ (‘The Brick’), distributed in mimeographed form and eventually printed in 1992.²⁷ This economic plan advocated for a gradual shift toward a market-based economy. In one of its chapters, El Ladrillo already contains the main features of the scheme based on individual capitalisation accounts, later known as the ‘AFP system’ after the for-profit Administradoras de Fondos de Pensiones (Pension Funds Administrators, AFPs) that managed workers’ savings. Prepared in the event of a military takeover, El Ladrillo largely defined the military regime's economic agenda and represented a compromise between the different segments of the broad business coalition that supported the coup.

The Pre-Coup Gradualist Business Coalition

Two segments of the economic elite played a leading role in this broad business coalition: (i) traditional, internationally competitive economic groups, and (ii) financial conglomerates with connections to international financial markets.²⁸ Differences in the conglomerates’ concentration in different sectors of the economy determined such categories. Given that the economic composition of business conglomerates is closely related to their interests (which, in turn, may determine distinct policy preferences and roles in the policy-making process), a brief analysis of their business portfolios is in order.²⁹ Conglomerates with a higher concentration in (highly mobile) liquid assets, for example, may push for faster, radical changes as they are able to adjust much more quickly than fixed-assets holders to market signals. Financial conglomerates held portfolios with a heavy concentration in liquid assets, with two-thirds or more of their companies in sectors such as banking, insurance, commerce and real estate speculation. In contrast, only roughly a half of the holdings of the traditional conglomerates were in liquid assets; the other half were in industrial, fixed-asset sectors (see Table 2). Importantly, traditional conglomerates constituted the dominant fraction in this broad, intersectoral gradualist coalition, while financial conglomerates occupied a subordinate position.³⁰

Table 2. Major Conglomerates in 1970 and Concentration of Companies by Economic Activity

Source: Adapted from Silva, The State and Capital in Chile.

Both groups favoured economic restructuring towards a market-based economy, including pension privatisation. Traditional conglomerates, the dominant segment, pushed for a gradual market-based restructuring with an emphasis on industrialisation: as they held interests in fixed-asset sectors, these groups could adapt only to a gradual, slower opening of the economy to external competition. By contrast, financial conglomerates – given their portfolios concentrated in liquid assets – preferred a swift and radical restructuring of the economy. The gradualist coalition, forged in the early 1970s during the political campaign against the Socialist presidential candidate Salvador Allende, was thus to be a fragile, short-lived alliance.³¹ By 1975, it would be superseded by a coalition led solely by the radical financial conglomerates (hereafter referred to as the ‘radical neoliberal coalition’ or simply ‘radical coalition’) and excluding traditional groups.³²

As different historical accounts attest, business opposition to the Allende government started soon after his election in 1970.³³ In mid 1971, top executives of the most influential traditional and financial conglomerates began to meet on a regular basis. Representatives of the Edwards group (traditional) and of the radical financial conglomerate BHC coalesced around what they called ‘The Monday Club’ (as they met every Monday for lunch). This network represented a key source of instrumental power, becoming the main locus of civilian organisation of the coup and coordination with the armed forces. From this network, the gradualist coalition was able to establish strong ties to the armed forces. These ties would allow the gradualist coalition to influence the economic agenda (and particularly the discussion on pensions) in the initial phase of the military regime.

One of the main participants in this network was ex-navy officer Hernán Cubillos, at that time top executive of the Edwards group, and future minister of Foreign Affairs during the dictatorship (1978–80). Cubillos constituted the chief link between the gradualist coalition and the armed forces. Navy officers who participated in the plot against Allende let it be known that they could not go ahead with the coup if they did not have a clear economic programme.³⁴ The communication channel for this message was via another ex-navy officer and top executive of the Edwards conglomerate, Roberto Kelly, who passed the message to Cubillos and would later play a key role recruiting neoliberal technocrats for the Oficina de Planificación Nacional (National Planning Office, ODEPLAN; see further below).³⁵

Months before the coup d’état, representatives of the Monday Club convened a group of top executives and Chicago Boys with the aim of drafting a post-coup economic programme: the afore-mentioned Ladrillo.³⁶ This policy document represented a compromise between the dominant, traditional conglomerates and the subordinate radical financial conglomerates. It advocated for economic restructuring, but proposed a gradual process without a defined calendar.³⁷

El Ladrillo's authors were neoliberal technocrats, materially connected with both traditional and financial conglomerates, who were to reach highly influential posts in the authoritarian regime's Cabinet. The lead author was Sergio de Castro, known as the ‘dean’ of the Chicago Boys. Linked to the Edwards group through his work for the Centro de Estudios Sociales (Social Studies Centre, CESEC), the think-tank funded by this conglomerate,³⁸ after the coup de Castro began his service in government as an advisor, becoming Minister of the Economy in April 1975. Another prominent author of El Ladrillo was Manuel Cruzat, a close friend of de Castro, who was top executive of the largest financial conglomerate, the BHC. In the early 1970s, Cruzat and his brother-in-law, Fernando Larraín, split off from the BHC group.³⁹ From 1974 onwards, then, Manuel Cruzat would lead the mammoth Cruzat-Larraín financial conglomerate.

Previous research has generally depicted the Chicago Boys as autonomous, failing to highlight their close contacts with the economic elite.⁴⁰ Top executives of economic conglomerates were closely allied to the radical Chicago Boys, forming a network that grew out of a shared social, professional and academic background. For instance, most of the Chicago Boys and some top executives had studied economics together at the Pontificia Universidad Católica and, in some cases, at the University of Chicago.⁴¹ Furthermore, the most prominent Chicago Boys were top executives of both traditional and financial conglomerates before and after their appointments to top positions in Pinochet's dictatorship.

As noted, El Ladrillo contains the first blueprint of the pension reform finally implemented in 1981. In one section, entitled ‘Welfare and Social Security Policies’, it introduces what was to constitute a paradigmatic shift: the transition from a pay-as-you-go pension plan to a fully funded scheme based on individual accounts.⁴² The new pension scheme would consist of two subsystems: a state-provided minimum pension and ‘the second, that would be based on a system of mandatory savings scheme that would be deposited monthly in a specialised institution … which savings could be drawn out only in the form of a life annuity on reaching retirement age’.⁴³ Given their significant holdings in liquid assets, pension privatisation as outlined in El Ladrillo suited the interests of both traditional and financial conglomerates. Among other opportunities provided by the new scheme, pension fund managers could invest in debt instruments issued by their banks and shares issued by their companies, and their insurance companies could sell life annuities to retirees.

Having coalesced around the Monday Club and offered a coherent economic plan, the gradualist business coalition was able to set the policy agenda early in the authoritarian government: the one detailed in El Ladrillo prevailed because it was a coherent programme. In contrast, the top military officers involved in the conspiracy did not have a clear economic plan.⁴⁴ Immediately after the coup some asked precisely those neoliberal technocrats who had worked on El Ladrillo for help. Thus, early on ties to the armed forces coupled with technical expertise constituted an additional source of instrumental power for the economic elite. In the aftermath of the coup, then, civilians materially connected to the gradualist coalition became ministers in influential positions, while the Chicago Boys were appointed as advisors.⁴⁵

Critical for the analysis of pension policy-making, the segment of internationally uncompetitive capitalists producing for the domestic markets constituted the only fraction of the economic elite which was not part of the broad, gradualist coalition just discussed. This segment did not favour a departure from the Import-Substitution Industrialisation (ISI) strategy and opposed a shift toward a market-based economy. These internationally uncompetitive industrialists had been passive supporters of the expansion of the social security system as it constituted a vital aspect of the bargain maintaining the multiclass alliance that supported the ISI model.⁴⁶ Nonetheless, Allende's policy of nationalisation had weakened this segment, thus reducing their influence.

Second Stage (September 1973–March 1976): Corporatist Officers Limit the Chances for Radical Pension Privatisation

The Outcome

During this second stage that extends from the military takeover (11 September 1973) to March 1976, the military Junta pursued a compromise pension bill. This bill attempted to harmonise the reform proposal developed in El Ladrillo with one prepared by high-ranking corporatist officers (the so-called ‘Anteproyecto’).⁴⁷ This group of corporatist officers – who were the main plotters of the military takeover – was independent of business power. As such, they pursued their own agenda. Proponents of El Ladrillo controlled ODEPLAN while corporatist military officers – who aimed to harmonise labour–capital relations as a means of purging Marxism from Chilean society and strengthening national security – were more influential within the Ministry of Labour and Social Security, formally in charge of the reform process. ODEPLAN was a key agency for the Chicago Boys: they controlled it immediately after the coup, and from there they designed and promoted their policies.⁴⁸

The compromise bill pursued by the government up to 1976 was, therefore, an attempt to harmonise contradictory, disparate concepts such as state subsidiarity (the principle of a limited role for the state being key for neoliberal technocrats) and universalism (supported by corporatist officers).⁴⁹ While ODEPLAN prevailed in the financial aspects of the proposal (i.e. creation of individual capitalisation accounts),⁵⁰ the Ministry of Labour prevailed in the determination of the administrative structure: pension funds were to be managed by semi-public, non-profit organisations run by workers to be called ‘Corporaciones de Seguridad Social’ (‘Social Security Corporations’). The Ministry of Labour officially presented the Anteproyecto to the public in November 1975, aiming to promote an open discussion. Nonetheless, the process came to an abrupt halt when the Chicago Boys took control of the Ministry of Labour in 1976 and blocked the proposal.

Pensions and the Gradualist Coalition

The dictatorship's initial ministerial structure reflected the balance of power within the economic elite. As we saw above, the broad, intersectoral gradualist coalition was dominated by traditional conglomerates, with financial groups playing a subordinate position. The radical neoliberal technocrats’ subordinate role in the government apparatus is evident in Figures 1a and 1b: in 1973 and 1974 Chicago Boys were in advisory positions and none of them occupied a ministerial post. Gradualists such as Fernando Léniz, former top executive of the Edwards group, occupied key positions like the Ministry of the Economy. The Ministry of Labour and Social Security was headed by a corporatist general, Nicanor Díaz, who was a very close associate of General Gustavo Leigh, member of the military Junta.

Figure 1. Evolution of the Composition of the Cabinet, 1973–81

Notes: Boxes shaded in grey indicate members of the Chicago Boys’ network.

DC Member of Democracia Cristiana

M Minister/Ministry

P President

SEGEGOB Secretaría General de Gobierno

VP Vice-president

Source: Authors’ elaboration based on Fontaine Aldunate, Los economistas y el presidente Pinochet; ‘Boris Blanco: la caída del superintendente’, Hoy, 15 March 1984, pp. 24–9; ‘Falleció ex canciller Hernán Cubillos’, El Mercurio, 11 April 2001, p. C1; Empresa Periodística de Chile, Diccionario Biográfico de Chile, 1984–1986, 18th edn (Santiago: Empresa Periodística de Chile, 1986); ‘Previsión: el debut de las AFP’, Hoy, 13 May 1981, pp. 30–3; ‘Quién es y cómo es Javier Vial, y el baile de los pesos’, Hoy, 26 Aug. 1981, pp. 22–6; ‘Quién es y cómo es Manuel Cruzat, el millonario invisible’, Hoy, 16 Sept. 1981, pp. 19–22; María Olivia Mönckeberg, El saqueo de los grupos económicos al Estado chileno (Santiago: Ediciones B, 2001); ‘Las once de la “competencia”’; ‘COPEC: los rostros nuevos’, Hoy, 24 June 1981, p. 43.

* Humberto Pizarro was an ‘outsider’: neither an economist nor a graduate of the Pontificia Universidad Católica.

ODEPLAN was led by Kelly, an ex-navy officer who opposed the high-ranking corporatist officers and was one of the main links between the military plotters and the economic elite. As mentioned above, after leaving the navy Kelly became top executive of the Edwards group, which was at the core of the gradualist coalition.⁵¹ He recruited a number of neoliberal technocrats to ODEPLAN. The most prominent was Miguel Kast, who played a critical role in the pension discussions from the beginning: in 1972, his report on the US system of private retirement accounts constituted the basis for the proposal developed in El Ladrillo.⁵² From then on Kast became the most prominent ideologist and promoter of the neoliberal reform of the pension system.⁵³

There were additional ties between the Chicago Boys and the armed forces. One of these links was Colonel Enrique Lackington, appointed Undersecretary for the Economy immediately after the coup. His son, also an economist, was a close friend of de Castro and recommended de Castro and other Chicago Boys as advisors to the Ministry of the Economy.⁵⁴ Ties to the armed forces, coupled with informal ties and recruitment of business representatives into government were, therefore, the most decisive sources of instrumental power for the gradualist coalition (and for financial conglomerates in particular) in the period following the military takeover.

The conflict between neoliberal economists and high-ranking corporatist officers had been acute since the beginning. Kelly, for instance, recalls that when Kast proposed in 1974 the replacement of a common fund with individual capitalisation accounts, ‘Leigh was furious, he said that it was madness to propose such reform.’⁵⁵ The economic elite pushed directly for neoliberal reform while they – and their Chicagoan associates – resisted the creation of workers’ associations to manage pension funds. Informal ties afforded the economic elite the opportunity to reach General Díaz himself and put pressure on him. Speaking to Hoy magazine in 1980, Díaz asserted that ‘many businessmen came directly to my office to criticise the bill given that, in their opinion, workers would not be able to manage their funds efficiently’.⁵⁶

For the authoritarian regime, the solution to this acute conflict was the compromise bill. The first draft of the Anteproyecto was formally submitted to the military Junta on 23 October 1974,⁵⁷ and the government announced the reform to the public in November 1975.⁵⁸ The proposal sought to solve the severe economic burden of the pay-as-you-go system by streamlining benefits. The bill also considered the introduction of a common fund and cross-subsidies in order to fund other social security programmes. The Ministry of Labour promoted open discussion by disseminating the reform in the media and carrying out dialogues with workers’ associations and specialists.⁵⁹ The result of this consultation process, which culminated in February 1976, was a four-volume report with comments and suggestions.⁶⁰

Yet this participatory process came to an abrupt halt in March 1976 when Sergio Fernández replaced General Díaz as the head of the Ministry of Labour. Fernández was a gremialista and hardliner, who had been working with Kast on ODEPLAN's reform proposal.⁶¹ Why did this change take place? Did it reflect a change in the balance of power within and outside the state?

Mutually Reinforcing Changes within the Economic Elite and the State

General Díaz had battled for the pre-eminence of corporatist ideas, opposing the Chicagoans’ uncompromising neoliberal orthodox views.⁶² Yet he grew increasingly isolated: the economic elite opposed his plans because they were too ‘statist’, while trade unions did not support them either. Díaz's resignation and the appointment of Fernández as Minister of Labour signalled the Chicago Boys’ growing ascendency and can be explained by mutually reinforcing changes within the economic elite and the state.

Disputes between traditional and financial conglomerates over the pace and extent of the economy's liberalisation broke the gradualist coalition in early 1975.⁶³ The short-lived gradualist coalition was to be replaced by the radical neoliberal coalition led by the financial conglomerates, which progressively enhanced their sources of structural and instrumental power.

A combination of international and domestic factors facilitated this process. At the beginning of the 1970s, only two conglomerates made up the financial segment's core (see Table 2): the Vial group and its spin-off, the Cruzat-Larraín conglomerate. Critically, these financial groups controlled the negotiation of foreign loans.⁶⁴ By late 1973, the first oil crisis, coupled with the lack of loans from multilateral lending banks, aggravated Chile's traditional foreign exchange bottlenecks. The country found itself without access to international financial institutions because of an international boycott of the authoritarian regime. Moreover, in 1974, a sharp fall in the price of copper – Chile's main export – made the external position even worse.

Financial conglomerates promoted the implementation of an orthodox stabilisation programme (so-called ‘shock therapy’) to solve the crisis. Breaking with the gradualist coalition, they pushed for drastic neoliberal economic measures including the deregulation of the financial sector, faster privatisation and tariff reductions. Crucially, in April 1975 Pinochet decided to support the implementation of the ‘shock therapy’ to stabilise the economy. This decision was followed by a Cabinet reshuffle, which crystallised the rise of the radical financial coalition to the top and meant the delegation of decision-making to the Chicago Boys. Firstly, Jorge Cauas, a former Christian Democrat and Chicago Boy sympathiser,⁶⁵ took over the post of ‘Super-Minister of Finance’ through a decree granting him extraordinary powers.⁶⁶ Secondly, de Castro replaced gradualist Fernando Léniz as Minister of the Economy.⁶⁷ By April 1975, then, neoliberal technocrats materially linked to financial conglomerates controlled all positions in the Economics Ministry (comparison of Figures 1b and 1c shows the speed and scale of the change). Chile thus began to embrace an extreme version of neoliberalism.⁶⁸

Pinochet's support for the shock therapy was decisive for its implementation: by late 1974, he had concentrated all executive power in his hands. A significant step in this direction came in mid 1974 with the creation of DINA (the military intelligence agency), which operated under Pinochet's direct command. He was also successful in gradually purging his most influential (mostly corporatist) political rivals within the army.⁶⁹

Why did Pinochet support the radical and risky stabilisation programme? The answer has little to do with economic theory: Pinochet did not have a clear policy agenda or much understanding of economics.⁷⁰ Rather, two other factors are more significant. First, according to Silva, Pinochet was well aware of the growing structural power of the radical financial coalition in the context of the severe credit crunch and hopeful that such groups would back neoliberal policies with increasing levels of investment.⁷¹ The fact that these conglomerates had the ability to provide foreign reserves made them very attractive to Pinochet.⁷² Second, Pinochet decided to support the Chicago Boys as a way of distancing himself from his adversaries inside the military Junta, especially the corporatist General Leigh.⁷³

The application of the shock treatment significantly strengthened the radical coalition. It gave financial conglomerates an advantageous position to acquire low-priced state companies during the recessionary period, while further weakening the internationally uncompetitive industrialists.⁷⁴ Sources of instrumental power were key for the further economic expansion of financial conglomerates, and, ultimately, for the enhancement of their structural power. Their relationships with the Chicago Boys afforded them critical insider information: neoliberal technocrats advised them of policy changes before they were announced to the public.⁷⁵ This access to privileged information allowed the Cruzat-Larraín and Vial conglomerates to expand rapidly.

A factor that further strengthened the radical business coalition was the expansion of its power base when the Edwards group, owner of the highly influential El Mercurio newspaper, joined the radicals. The Edwards group, with its sources of instrumental power (e.g. media access), contributed to the swift, radical restructuring of Chile's economy. In November 1974, for example, El Mercurio published its first editorial advocating rapid privatisation, a reduction in tariffs and drastic deflationary policies.⁷⁶ By early 1975 three powerful conglomerates comprised the core of the radical coalition: Cruzat-Larraín, Vial and Edwards, which were to throw all their weight behind the privatisation of the pension system.

The increasing centralisation of power of the authoritarian regime thus strengthened the radical business coalition. The establishment of one-man rule, without parallel within contemporary Latin American authoritarian regimes,⁷⁷ was pivotal for the neoliberal project.⁷⁸ With such power, Pinochet was to insulate the Chicago Boys from the pressure of those groups in a society negatively affected by the radical economic restructuring.⁷⁹ The opposition of the neoliberal coalition to the corporatist-led reform proposal was ultimately successful: according to the minutes of its secret sessions, after General Díaz's resignation the Junta never discussed the Anteproyecto again.

Third Stage (March 1976–May 1981): Imposition of the Private Pension System

The Outcome

The reform finally implemented in November 1980 (Law no. 3500)⁸⁰ put the management of savings for pensions in the hands of for-profit corporations, which had been a central goal of neoliberal technocrats since the beginning.⁸¹ The proposal legally mandated each worker to transfer 10 per cent of her/his taxable income to an individual savings account managed by AFPs. The AFPs’ profits were to arise mainly from a fixed commission – established with a flat, regressive structure – charged to every worker. Critically, the reform did not include mechanisms for workers’ participation in the management of their savings. Under the new system, moreover, workers would receive a pension based solely on their savings plus returns (employers would not contribute to retirement accounts). This structure ultimately meant that a growing stream of savings was to be managed by for-profit providers established by financial conglomerates.

During this period, financial business groups dominated the economic elite, having displaced traditional economic groups from the top. This change in the economic elite's balance of power triggered further changes in the composition of the Cabinet, which became firmly dominated by neoliberal technocrats materially connected to the largest financial conglomerates. Accordingly, during this period we observe a highly exclusionary policy process characterised by unmediated access of the economic elite's financial fraction to the centres of policy-making.

Radical Neoliberal Coalition's Further Accumulation of Power, 1976–8

After the application of the shock treatment in 1975, the radical financial coalition continued expanding its economic base and, therefore, its sources of structural power.⁸² By 1978, the Cruzat-Larraín and Vial conglomerates together controlled 48.3 per cent of private banking assets and 26 per cent of the country's external credit, increasing to 52 per cent by 1981.⁸³ Also by 1978, these two conglomerates alone controlled more than 37 per cent of the assets of Chile's largest 250 companies.⁸⁴

During this period, additional economic groups joined the radical business coalition. Several traditional conglomerates restructured their business strategies in order to take advantage of the opportunities created by the neoliberal reforms. As we have already seen, Edwards was the first conglomerate to pursue this strategy.⁸⁵ The powerful traditional Matte and Luksic groups followed suit in subsequent years.

As the radical coalition's control of the economy and its sources of structural power grew, their sources of instrumental power also increased. Critically, the radical coalition built closer relationships with policy-makers by, for example, the recruitment into government of more of its members. The paradigmatic case is the appointment of José Piñera as Minister of Labour and Social Security in December 1978. Piñera embodied the direct access of the radical financial coalition to the policy-making process. Before heading the Ministry, he was a top executive of the Cruzat-Larraín conglomerate advocating for the adoption of ODEPLAN's original radical proposal. Working in close collaboration with Cruzat, Piñera waged a campaign for the imposition of private accounts managed by private providers from 1976. The campaign benefited from the support of newspapers controlled by the Edwards conglomerate, which played a crucial role in savaging the pay-as-you-go system (see more on this below) and legitimising radical reform.⁸⁶ Privileged media access was another source of instrumental power greatly strengthened during this period. The Cruzat-Larraín conglomerate, for instance, acquired the influential Ercilla magazine in September 1976.⁸⁷ Piñera regularly wrote op-eds for the magazine, as did Kast.⁸⁸

The Final Road to Privatisation, 1979–81

Notwithstanding their growing clout, neoliberals could not impose their project between 1976 and 1978. According to General Díaz, ‘they [the Chicago Boys] did not dare to touch the reform for two years after I left the ministry’.⁸⁹ The main explanation for the deadlock is Leigh's continuing role in the Junta. The turning point came only in July 1978, when Pinochet removed General Leigh from the Junta, thus giving further impetus to the pension reform.⁹⁰ Leigh was the only member of the military Junta with connections to the segment of internationally uncompetitive industrialists (excluded from both the gradualist and radical coalitions) and his departure made the policy process even more one-sided.⁹¹

After Leigh's departure the Cabinet became totally dominated by neoliberal technocrats, all members of the radical coalition. Five of the 12 top government positions shown in Figure 1d were held by top executives, employees or board members of the Cruzat-Larraín, Vial and Edwards groups. Five others (Fernández, Juan Carlos Méndez, Sergio de la Cuadra, Kast and Alfonso Serrano) did not have direct employment or business links with conglomerates at the time but were committed neoliberals; the last two (Generals René Vidal and Carlos Jiménez) had close ties to Pinochet.

After Piñera's appointment to the Ministry of Labour in December 1978, he convened a commission charged with detailing the reform.⁹² The commission, chaired by Martín Costabal (a former top executive of the Matte group), began to meet weekly⁹³ and held regular meetings with representatives of the Ministry of Finance.⁹⁴ It worked on actuarial studies and the specific reform design following ODEPLAN's earlier ideas.⁹⁵ Later, in early 1980, Pinochet convened a joint, special legislative commission charged with the single task of discussing the proposal prepared by Piñera's team.⁹⁶

Publicly, the first announcement of the reform came in May 1980. In the May Day commemoration, the government officially declared 1980 ‘the year of social security’.⁹⁷ At that event, both Pinochet and Piñera declared to an audience of 3000 attendees that ‘the old system was totally bankrupt’.⁹⁸ As noted above, the campaign for the neoliberal reform was chiefly based on savaging the pay-as-you-go system, which entailed disseminating inaccurate facts and figures. During our research we gained access to a secret, unpublished report prepared by the Ministry of Finance's Budget Office, which neoliberal technocrats used to support their claims for the alleged bankruptcy of the pay-as-you-go system.⁹⁹ The document constitutes the main historical proof of the blatant, deliberate attempt to discredit the old pay-as-you-go system and to legitimise the radical pension reform promoted by the Chicago Boys.

The argument developed in the secret report lacks scientific rigour. For instance, in computing the overall financial balance of the system, it makes the indefensible assumption that every Chilean worker would automatically switch from the pay-as-you-go scheme to the AFP system once the latter was implemented. This situation would inevitably mean that the pay-as-you-go scheme would instantly have been seriously underfunded. Following this assumption, the authors underestimated the assets of the old system since they did not include the present value of workers’ payroll contributions.

The legislative process slowed for some months while the government concentrated on the constitutional referendum scheduled for September 1980.¹⁰⁰ Then, Piñera cunningly seized the opportunity offered by the triumphant atmosphere in the aftermath of the referendum to convince Pinochet to discuss the bill in the Junta's secret sessions.¹⁰¹ The Junta discussed and passed the reform over two sessions, 14–16 and 24 October.¹⁰² As mentioned in the introduction above, Pinochet raised his concerns regarding the management of the funds in the opening of the first session: ‘I also agree on the fact that the social security system has to be modified … But at the same time, I have said that I do not agree with the idea that the funds go to the private sector.’¹⁰³ Then, he openly proposed that funds be managed by the state:

Therefore, Minister [Piñera], everything is OK, but when we discuss the issue of the management of the funds, I hesitate. Let me explain: could it be that one of our agencies, a responsible state agency … such as the Central Bank, that such an agency could hold the money and assign it to banks and ‘cajas’, I don't know, I can't work it out at the moment.¹⁰⁴

Later, he once again made clear his reservations about the role the private sector was to play in the defined-contribution, for-profit system:

I maintain the apprehensions I pointed out before. Why? Because here there are two gentlemen who are trying to grab … I will not name them, but you know perfectly well who they are since Hoy magazine published a report on this … The machinery is so big, but if you remove the nut at the bottom the whole building collapses. The evidence is clear: they have built an empire. Therefore, I have found out … they will be those who will ruin the [pension] system.¹⁰⁵

According to Carlos Huneeus, ‘the two gentlemen’ whom Pinochet was specifically pointing at in this passage were the heads of the two largest financial conglomerates, Cruzat, Minister Piñera's former boss, and Vial, head of the Vial group.¹⁰⁶ Pinochet was thus clearly expressing his fears that financial conglomerates could exploit the privatised pension system to their benefit. From information provided by the press, by the Comité Asesor de la Junta (Advisory Committee to the Junta, COAJ) and by military intelligence, Pinochet was very aware of the structure of the financial conglomerates and the way in which they operated.¹⁰⁷

Pinochet expressed his concerns 14 times during this secret session. Why, then, in the end did he support the radical pension reform? The answer lies to a large extent in the radical coalition's structural power augmented by their instrumental power: as we have seen, by 1980 financial conglomerates controlled vast sectors of Chile's economy, and Pinochet needed their investment in order to foster economic growth and consolidate his rule.

Critically, in the days before the Junta's secret sessions El Mercurio waged a campaign stressing the urgent need to boost investment. Just two days after the constitutional referendum, on 13 September 1980, the conservative daily urged ‘the authorities’ [i.e. Pinochet] in its editorial page to formulate a development programme that would ‘further clarify the rules of the game and propose specific policies in critical aspects for development with stability, with special emphasis on issues not yet resolved, such as the size of the state, the structure of public spending, the improvement of the investment rate … and the extension of property [ownership] to all Chileans’.¹⁰⁸ El Mercurio called for a reduction in the size of the state and an expansion of the role of the private sector in the economy.

In the same editorial, the newspaper announced ‘the upcoming pension reform’ and its potential effects (even before the Junta had discussed the reform proposal). Under the sub-heading ‘Investment must increase’, El Mercurio insisted that ‘private development in areas such as social security … is, in practice, hindered by the existing quasi-monopoly of the state’. The editorial also advocated for the elimination of employers’ contributions to the social security fund, which was incorporated to the letter in the reform approved by Pinochet: ‘Social security tax still exists and it seems essential to eliminate it completely in the coming years.’

As the discussion of the reform in the Junta drew nearer, pressure became even more explicit. For example, on 28 September 1980, under the sub-heading ‘Progress that Cannot be Delayed’, El Mercurio's editorial stated: ‘It is inevitable that the Government must keep moving forward relentlessly to implement the programme it has formulated.’ The editorial then listed the ‘modernisations’ proposed by the Chilean government, which were intended to ‘liberalise’ individuals and society. The first was the Plan Laboral of 1979, which, among other things, restricted the right to strike.¹⁰⁹ ‘The second modernisation … is the reform to social security and, specifically, to the old-age pension system.’¹¹⁰ Stridently, under the sub-heading ‘The Importance of the Pension Reform’, the newspaper continued to stress the topic by asserting: ‘The reform means a mandatory allocation of substantial funds to productive investment.’ This campaign thus represents a textbook case of structural power augmented by instrumental power:¹¹¹ the country's most influential newspaper, controlled by the radical financial coalition, used its editorial pages, the ones which set the daily political agenda at that time, to stress time and again the urgent need for pension privatisation.

In October discussion in the Junta became fraught with military mistrust, on the one hand, and the Chicago Boys’ insistence on the alleged benefits of the reform, on the other. Piñera's attempts to convince a mistrustful military leadership were carefully framed. On the second day of the first secret session, held on 16 October 1980, he highlighted the importance of creating workers committed to private ownership:

By having an individual account and seeing that his savings rise, which depends on how the economy is doing, the worker will necessarily be interested in having an efficient and responsible Minister of Finance, in having Parliaments that are not so committed to partisan politicking, and in avoiding violent revolutions, because he will become a property owner with a direct interest in the general performance of the economy. In this way, there will be fewer strikes.¹¹²

The few high-ranking officers that still defended the corporatist agenda then raised additional concerns. For instance, Major Juan Romero, Head of Legislation for the COAJ, objected that the alleged bankruptcy of the old system reported by Piñera was based on dubious accounting criteria.¹¹³ Yet most members of the Junta readily bought in to Piñera's overstated arguments; their intimidation at the hands of the Chicago Boys when they tried to inquire about further details is apparent in the minutes.¹¹⁴ Piñera grossly misled the Junta during the debate. For example, he said: ‘I want to emphasise what I have said before: the change … will mean zero cost to the state.’¹¹⁵ The statement soon proved inaccurate: the state had to continue paying pensions to retirees of the old system but simultaneously received far lower contributions from active workers, as the number of affiliates decreased from ca. 1.7 million in 1980 to just below 450,000 by 1984.¹¹⁶ Alberto Arenas reports that the reform caused a total budget deficit of 5.7 per cent of GDP from 1981 to 1998, more than doubling the deficit for the 1974–80 period.¹¹⁷

The reform was finally approved in early November 1980. The subsequent implementation stage of the policy-making process offers a further example of the power of the business elite. Their unmediated access to policy-making provided financial conglomerates with privileged information even before the final approval. Hoy magazine reported that from 15 October to 4 November 1980, Chile's government gazette published the registration of 57 original trademarks that might be used for the new AFPs.¹¹⁸ These names were registered before the new pension system was announced to the public on 6 November 1980.

The launch of the AFP system was a success. Most of the economically active population moved to the new system during the first six months.¹¹⁹ The most powerful financial groups became key actors in the new system. Of the 11 AFPs initially created,¹²⁰ the Cruzat-Larraín conglomerate controlled two (ProVida and Alameda), the Vial Group another two (Santa María and San Cristóbal), and the Edwards conglomerate controlled El Libertador. Those traditional groups which had successfully adapted to the changing environment in the previous three years also benefited: the Luksic, Matte and Angelini conglomerates created Summa AFP, and the Yarur-Banna and Said conglomerates joined together to create PlanVital. The Cámara Chilena de la Construcción (Chilean Chamber of Construction, CChC), in turn, created Habitat AFP. In summary, ten out of 11 corporations were owned either by conglomerates or by big business interests.

Furthermore, the cosy relationship between radical conglomerates and the Chicago Boys was reflected in the structure of the boards. For instance, the chairman of Provida was Cauas, former Minister of Finance, and Alfonso Márquez de la Plata, former Minister of Agriculture, was a member of the board. Invierta counted Álvaro Bardón, former President of the Central Bank, as board chairman. The renown of these economists – along with the prestige of the conglomerates behind the AFPs – were presented as the guarantee of the efficiency, success and ‘modernity’ of the system.

Conclusions

This paper has illustrated the suitability of the instrumental–structural framework for the analysis of business power under authoritarian regimes. By adapting the instrumental–structural framework to non-democratic settings, we show that these categories are useful to account for the economic elite's progressive accumulation of power in authoritarian Chile. More specifically, the financial segment progressively accumulated sources of power until it was able to secure its goal of privatising the old-age pension system.

Multiple sources of power enabled financial conglomerates to defeat their opponents, while its state allies simultaneously gained more influence. The unmediated, privileged interaction between the state and the financial fraction of the economic elite gradually increased, reaching its zenith between 1979 and 1981, when neoliberal civilians dominated the Cabinet. This resulted in the creation of the new pension scheme based on individual accounts managed by for-profit companies controlled by the financial conglomerates. Even General Augusto Pinochet – arguably the most powerful dictator in Latin America during the 1960s and 1970s – ended up accepting a reform that he did not particularly like.

In addition to the contributions that we make in this work to an understanding of business power in non-democratic contexts, we offer a more nuanced view than does previous literature: the state is not a transmission belt of business interests, nor is it completely autonomous. Policy-making in authoritarian regimes, then, results from a complex interplay between dynamics within both the state and the economic elite. These dynamics, in turn, are influenced by the power resources wielded by each actor.

Overall, in rethinking the policy-making process in authoritarian contexts we make a broader contribution to the literature on the political economy of business–state relations in Latin America and beyond. We have shown that – important as it is – the study of the regime's structure is not enough for the formulation of an accurate picture of social policy reform in authoritarian contexts in general and Pinochet's Chile in particular. Likewise, factors that have previously occupied a central role in explaining outcomes – such as the ideological position of policy-makers – fall short of providing a sound explanation. Instead, we stress the need to include in the analysis the power resources of the different actors involved in the policy process, and the study of their material interests and relations. Given its centrality in shaping public policies in unequal contexts, Latin Americanists and other scholars of the global South should pay special attention to the role played by economic elites and their different fractions, in both the past and the future.